Modulation Prediction - CNN Configuration 1¶

Setting up Training Data¶

In this section, we are importing the training and testing test set in .hdf5 format and converting them to numpy arrays.

import numpy as np
import h5py
import csv
import matplotlib.pyplot as plt
with h5py.File("data.hdf5", 'r') as f:
    train_data = f['train'][()]
    test_data = f['test'][()]
labels = open('train_labels.csv')
labels_reader = csv.reader(labels)
classes = ["FM","OQPSK","BPSK","8PSK","AM-SSB-SC","4ASK","16PSK","AM-DSB-SC","QPSK","OOK"]
train_labels_data = []
for row in labels_reader:
    if row[0] != 'Id':
        train_labels_data.append(classes.index(row[1]))

train_labels_data = np.asarray(train_labels_data)
print(train_data.shape)
print(len(train_labels_data))
print(test_data.shape)

trainlabelfft = train_labels_data

(30000, 1024, 2)
30000
(20000, 1024, 2)

Data Augmentation¶

In this section, we are going to use data augmentation as a technique for preprocessing the data. 4000 training samples were augmented with a rotation angle of 90 degrees.

theta = np.pi/2
counter = 0
for i in train_data:
    print(counter)
    if counter > 4000:
        break
    # rotation matrix
    rot = np.array([[np.cos(theta), -1*np.sin(theta)],[np.sin(theta),np.cos(theta)]])
    np.append(train_data,np.dot(rot,i.transpose()).transpose())
    np.append(train_labels_data,train_labels_data[counter])
    counter+=1
    
train_data

array([[[ 0.81229824,  0.35837647],
        [ 1.5064487 ,  0.32473826],
        [-1.3345141 , -0.11483677],
        ...,
        [ 0.4622633 ,  0.00232144],
        [-0.44701815,  0.3409667 ],
        [ 0.0929121 , -1.250979  ]],

       [[ 0.8651555 ,  1.1208268 ],
        [ 0.93955   ,  1.0812321 ],
        [ 1.0046352 ,  0.9699535 ],
        ...,
        [-0.17610398, -0.0531827 ],
        [-0.0274519 ,  0.10420295],
        [ 0.05284544,  0.23353934]],

       [[-0.27755857,  0.47558302],
        [ 0.269747  , -0.36386353],
        [ 0.7656934 , -0.3021099 ],
        ...,
        [-0.44257054,  0.25459397],
        [-0.93191254, -0.61050004],
        [ 0.5299933 ,  0.09297071]],

       ...,

       [[-1.2948849 ,  0.8575006 ],
        [-0.52819824, -0.70112824],
        [-0.8181306 ,  0.5288491 ],
        ...,
        [-0.34768716,  0.4826538 ],
        [ 0.14753036,  0.7205618 ],
        [-0.7184609 , -0.23059411]],

       [[-0.20838957,  0.28949115],
        [-1.2750533 ,  0.2082836 ],
        [-0.995211  ,  0.16588576],
        ...,
        [ 0.4434878 , -0.6900661 ],
        [-0.05305082,  0.05295856],
        [ 0.14935887,  0.1882925 ]],

       [[-0.2306667 , -0.42676774],
        [ 0.9506056 , -0.09998692],
        [-0.02584958,  0.3617045 ],
        ...,
        [-0.9461389 ,  0.84369624],
        [ 0.6568758 ,  0.04648934],
        [-0.66253257,  0.1033797 ]]], dtype=float32)

Visualizing Data¶

In order to get a sense for the "patterns" of the different labels, we will plot the first 100 samples from the training set. The X-axis will be the interphase component while the quadrature component is on the Y-axis. Each training sample has 1024 points with I/Q coordinates.

for j in range(0,100):
    I = []
    Q = []
    for i in range(0,1024):
        I.append(train_data[j][i][0])
        Q.append(train_data[j][i][1])
        
    plt.figure(j, figsize=(12, 8))
    plt.scatter(I,Q, label=classes[train_labels_data[j]])
    plt.title(classes[train_labels_data[j]])
    plt.xlabel('I')
    plt.ylabel('Q')
    plt.legend(loc='best')
    plt.grid()
    plt.show()

FFT Data Generation¶

In this section, we will generate the Fast-Fourier Transform of the I/Q signals. Each sample will still be 1024 x 2 and will carry the complex and real components. I experimented with this FFT input format for my CNN architecture. However, I found it to be inferior than the I/Q input format. Thus, for the competition, I did not use FFT for generating labels for the test set.

train_data_cmplx = train_data[:,:,0] + 1j*train_data[:,:,1]
fft_train_data = np.fft.fft(train_data_cmplx,axis=1)
fft_re = fft_train_data.real
fft_im = fft_train_data.imag

fft_proc = np.zeros(shape=train_data.shape)
for i in range(len(fft_proc)):
    fft_proc[i] = np.stack((fft_re[i],fft_im[i]),axis=1)
assert(fft_proc.shape == train_data.shape)

Training and Testing Splitting¶

In this section, we are converting the training and testing sets into Torch tensor format. After that I have split the training set into several smaller segments. I did this because I wanted to experiment with the batch size and analyze how it affects overfitting/underfitting. In addition, a validation set was created in order to see if the model is overfitting/underfitting during training.

from __future__ import print_function
import torch
from torch.nn import Linear, ReLU, CrossEntropyLoss, Sequential, Conv2d, MaxPool2d, Module, Softmax, BatchNorm2d, Dropout
from torch.optim import Adam, SGD
from torch.autograd import Variable
import torch.nn.functional as F
from sklearn.metrics import accuracy_score
from sklearn.model_selection import train_test_split
import torchvision
from torchvision import datasets, transforms

# converting training images into torch format
train_x = train_data.reshape(30000, 1, 1024, 2)
train_x  = torch.from_numpy(train_x)

# fft_proc = fft_proc.reshape(30000, 1, 1024, 2)
# fft_proc  = torch.from_numpy(fft_proc)

# converting test images into torch format
test_x = test_data.reshape(20000, 1, 1024, 2)
test_x  = torch.from_numpy(test_x)

# converting the target into torch format
train_y = train_labels_data.astype(int)
train_y = torch.from_numpy(train_y)
# trainlabelfft = trainlabelfft.astype(int)
# trainlabelfft = torch.from_numpy(trainlabelfft)

# split into batches
train_x, val_x, train_y, val_y = train_test_split(train_x, train_y, test_size = 0.4,shuffle=True)
# trainfftx, valfftx, trainffty, valffty = train_test_split(fft_proc, trainlabelfft, test_size = 0.4,shuffle=True)


train_x1, train_x2, train_y1,train_y2 = train_test_split(train_x, train_y, test_size = 0.50,shuffle=True)
train_x3, train_x4, train_y3,train_y4 = train_test_split(train_x1, train_y1, test_size = 0.50,shuffle=True)
train_x5, train_x6, train_y5,train_y6 = train_test_split(train_x2, train_y2, test_size = 0.7,shuffle=True)
train_val1, train_val2, train_valy1,train_valy2 = train_test_split(val_x, val_y, test_size = 0.95,shuffle=True)
# batches
train = torch.utils.data.TensorDataset(train_x,train_y)
# trainfft = torch.utils.data.TensorDataset(trainfftx,trainffty)
trainloader = torch.utils.data.DataLoader(train, batch_size=240, shuffle=True)
# trainloaderfft = torch.utils.data.DataLoader(trainfft, batch_size=240, shuffle=True)

# shape of training data
print(train_x.shape)
print(train_y.shape)

torch.Size([18000, 1, 1024, 2])
torch.Size([18000])

Setting up Convolutional Neural Network Architecture¶

In this section, we are specifying the Convolutional Neural Network Architecture for classification. For the framework, I have decided to use PyTorch, as it is relatively lightweight. This specific Jupyter Notebook is for one of two configurations of my CNN. In this configuration, I have used 2 convolution/pooling layers, followed by 3 fully-connected layers. In addition, I have used batch normalization for the first convolution/pooling layer to reduce covariate shift within the network as well as dropout to reduce overfitting. SELU activation were used.

Convolution layers: (1) 50 filters, 3X3 kernel size, Stride of 1, Padding of 3 (2) 30 filters, 5x5 kernel size, Stide of 1, Padding of 2

Pooling: Max Pooling with kernel size of 2X2

Fully-connected: (1) 7710 neurons to 128 neurons (2) 128 to 32 neurons (3) 32 to 10 neurons (output layer)

import torch.nn as nn

class Net1(nn.Module):
    def __init__(self):
        super(Net1, self).__init__()
        self.conv1 = nn.Conv2d(1, 50, kernel_size=(3,3), stride=1, padding=3)
        self.pool = nn.MaxPool2d(kernel_size=(2,2))
        self.batchnorm1 = nn.BatchNorm2d(num_features=50)
        self.conv2 = nn.Conv2d(50, 30, kernel_size=5, stride=1, padding=2)
        self.fc1 = nn.Linear(7710, 128)
        self.fc2 = nn.Linear(128, 32)
        self.fc3 = nn.Linear(32, len(classes))
        self.drop = nn.Dropout(p=0.3, inplace=False)

    def forward(self, x):
        x = self.pool(F.selu(self.batchnorm1(self.conv1(x))))
        x = self.pool(F.selu((self.conv2(x))))
        x = x.view(-1, 7710)
        x = F.selu((self.drop(self.fc1(x))))
        x = F.selu((self.fc2(x)))
        x = self.fc3(x)
        return F.log_softmax(x, dim=1)
        
cnn = Net1()
print(cnn)

it = iter(trainloader)
#itfft = iter(trainloaderfft)
#calculate dimensions
x = torch.randn(3,1,1024,2)
conv1 = nn.Conv2d(1, 100, kernel_size=(3,3), stride=1, padding=3)
y=conv1(x)
print(y.shape)
pool = nn.MaxPool2d(kernel_size=(2,2))
y=pool(y)
print(y.shape)
conv2 = nn.Conv2d(100, 30, kernel_size=5, stride=1, padding=2)
y=conv2(y)
print(y.shape)
y=pool(y)
print(y.shape)

Net1(
  (conv1): Conv2d(1, 50, kernel_size=(3, 3), stride=(1, 1), padding=(3, 3))
  (pool): MaxPool2d(kernel_size=(2, 2), stride=(2, 2), padding=0, dilation=1, ceil_mode=False)
  (batchnorm1): BatchNorm2d(50, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (conv2): Conv2d(50, 30, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2))
  (fc1): Linear(in_features=7710, out_features=128, bias=True)
  (fc2): Linear(in_features=128, out_features=32, bias=True)
  (fc3): Linear(in_features=32, out_features=10, bias=True)
  (drop): Dropout(p=0.3, inplace=False)
)
torch.Size([3, 100, 1028, 6])
torch.Size([3, 100, 514, 3])
torch.Size([3, 30, 514, 3])
torch.Size([3, 30, 257, 1])

Set up training protocol¶

Here, we are setting up the mini-batch training protocol. A batch size of 240 was chosen. The Adam optimizer was chosen and a learning rate of 0.0002 was tuned. In total, the max epoch was set to 80 and the cross entropy loss was selected. Every 20 batch iterations, the validation accuracy is printed.

def fit(model, train_loader,trainacc,testacc,trainloss):
    optimizer = torch.optim.Adam(model.parameters(), lr=0.0002)
    error = nn.CrossEntropyLoss()
    EPOCHS = 80
    model.train()
    for epoch in range(EPOCHS):
        for batch_idx, (X_batch, y_batch) in enumerate(train_loader):
            var_X_batch = Variable(X_batch).float()
            var_y_batch = Variable(y_batch)
            optimizer.zero_grad()
            output = model(var_X_batch)
            loss = error(output, var_y_batch)
            loss.backward()
            optimizer.step()
            predicted = torch.max(output.data, 1)[1] 
            print("Accuracy:")
            print(((predicted == var_y_batch).sum()).double()/240)
            trainacc.append((((predicted == var_y_batch).sum()).double()/240))
            print("Loss:")
            print(loss)
            trainloss.append(loss)
            print("Epoch:")
            print(epoch)
            print("Batch:")
            print(batch_idx)
            if (batch_idx % 20 == 0):
                with torch.no_grad():
                    validout = model(train_val1.float())
                valpred = torch.max(validout.data, 1)[1] 
                print("########################")
                print("Validation Accuracy:")
                print(((valpred == train_valy1).sum()).double()/len(train_valy1))
                testacc.append((((valpred == train_valy1).sum()).double()/len(train_valy1)))
                print("########################")

trainacc,testacc,trainlossainloss = [], [], []
fit(cnn,trainloader,trainacc,testacc,trainlossainloss)

Accuracy:
tensor(0.1250, dtype=torch.float64)
Loss:
tensor(2.3176, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
0
########################
Validation Accuracy:
tensor(0.0967, dtype=torch.float64)
########################
Accuracy:
tensor(0.0750, dtype=torch.float64)
Loss:
tensor(2.4866, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
1
Accuracy:
tensor(0.0750, dtype=torch.float64)
Loss:
tensor(2.4061, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
2
Accuracy:
tensor(0.1000, dtype=torch.float64)
Loss:
tensor(2.3322, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
3
Accuracy:
tensor(0.0667, dtype=torch.float64)
Loss:
tensor(2.4044, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
4
Accuracy:
tensor(0.0958, dtype=torch.float64)
Loss:
tensor(2.3356, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
5
Accuracy:
tensor(0.1333, dtype=torch.float64)
Loss:
tensor(2.3166, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
6
Accuracy:
tensor(0.1417, dtype=torch.float64)
Loss:
tensor(2.3364, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
7
Accuracy:
tensor(0.1000, dtype=torch.float64)
Loss:
tensor(2.3418, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
8
Accuracy:
tensor(0.1042, dtype=torch.float64)
Loss:
tensor(2.3049, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
9
Accuracy:
tensor(0.1583, dtype=torch.float64)
Loss:
tensor(2.2775, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
10
Accuracy:
tensor(0.1208, dtype=torch.float64)
Loss:
tensor(2.3103, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
11
Accuracy:
tensor(0.1667, dtype=torch.float64)
Loss:
tensor(2.2872, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
12
Accuracy:
tensor(0.1333, dtype=torch.float64)
Loss:
tensor(2.2800, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
13
Accuracy:
tensor(0.1417, dtype=torch.float64)
Loss:
tensor(2.2642, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
14
Accuracy:
tensor(0.1292, dtype=torch.float64)
Loss:
tensor(2.2707, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
15
Accuracy:
tensor(0.1667, dtype=torch.float64)
Loss:
tensor(2.2795, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
16
Accuracy:
tensor(0.1208, dtype=torch.float64)
Loss:
tensor(2.2984, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
17
Accuracy:
tensor(0.1500, dtype=torch.float64)
Loss:
tensor(2.2593, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
18
Accuracy:
tensor(0.1625, dtype=torch.float64)
Loss:
tensor(2.2753, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
19
Accuracy:
tensor(0.1458, dtype=torch.float64)
Loss:
tensor(2.2451, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
20
########################
Validation Accuracy:
tensor(0.1767, dtype=torch.float64)
########################
Accuracy:
tensor(0.1542, dtype=torch.float64)
Loss:
tensor(2.2562, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
21
Accuracy:
tensor(0.1583, dtype=torch.float64)
Loss:
tensor(2.2662, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
22
Accuracy:
tensor(0.1833, dtype=torch.float64)
Loss:
tensor(2.2016, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
23
Accuracy:
tensor(0.1583, dtype=torch.float64)
Loss:
tensor(2.2065, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
24
Accuracy:
tensor(0.2125, dtype=torch.float64)
Loss:
tensor(2.1773, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
25
Accuracy:
tensor(0.1542, dtype=torch.float64)
Loss:
tensor(2.2117, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
26
Accuracy:
tensor(0.1750, dtype=torch.float64)
Loss:
tensor(2.2244, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
27
Accuracy:
tensor(0.1917, dtype=torch.float64)
Loss:
tensor(2.1851, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
28
Accuracy:
tensor(0.1500, dtype=torch.float64)
Loss:
tensor(2.2416, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
29
Accuracy:
tensor(0.1833, dtype=torch.float64)
Loss:
tensor(2.1919, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
30
Accuracy:
tensor(0.1875, dtype=torch.float64)
Loss:
tensor(2.2349, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
31
Accuracy:
tensor(0.1708, dtype=torch.float64)
Loss:
tensor(2.2318, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
32
Accuracy:
tensor(0.1583, dtype=torch.float64)
Loss:
tensor(2.2069, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
33
Accuracy:
tensor(0.2042, dtype=torch.float64)
Loss:
tensor(2.1354, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
34
Accuracy:
tensor(0.1958, dtype=torch.float64)
Loss:
tensor(2.1738, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
35
Accuracy:
tensor(0.1625, dtype=torch.float64)
Loss:
tensor(2.1603, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
36
Accuracy:
tensor(0.1792, dtype=torch.float64)
Loss:
tensor(2.1875, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
37
Accuracy:
tensor(0.1875, dtype=torch.float64)
Loss:
tensor(2.1690, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
38
Accuracy:
tensor(0.2000, dtype=torch.float64)
Loss:
tensor(2.1673, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
39
Accuracy:
tensor(0.1583, dtype=torch.float64)
Loss:
tensor(2.1856, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
40
########################
Validation Accuracy:
tensor(0.1517, dtype=torch.float64)
########################
Accuracy:
tensor(0.2042, dtype=torch.float64)
Loss:
tensor(2.1501, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
41
Accuracy:
tensor(0.1625, dtype=torch.float64)
Loss:
tensor(2.1868, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
42
Accuracy:
tensor(0.1542, dtype=torch.float64)
Loss:
tensor(2.1626, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
43
Accuracy:
tensor(0.1708, dtype=torch.float64)
Loss:
tensor(2.1925, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
44
Accuracy:
tensor(0.1333, dtype=torch.float64)
Loss:
tensor(2.1749, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
45
Accuracy:
tensor(0.2042, dtype=torch.float64)
Loss:
tensor(2.1417, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
46
Accuracy:
tensor(0.1958, dtype=torch.float64)
Loss:
tensor(2.1232, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
47
Accuracy:
tensor(0.2250, dtype=torch.float64)
Loss:
tensor(2.1260, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
48
Accuracy:
tensor(0.2250, dtype=torch.float64)
Loss:
tensor(2.1175, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
49
Accuracy:
tensor(0.2125, dtype=torch.float64)
Loss:
tensor(2.1200, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
50
Accuracy:
tensor(0.2333, dtype=torch.float64)
Loss:
tensor(2.0355, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
51
Accuracy:
tensor(0.1708, dtype=torch.float64)
Loss:
tensor(2.1693, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
52
Accuracy:
tensor(0.2042, dtype=torch.float64)
Loss:
tensor(2.1183, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
53
Accuracy:
tensor(0.2333, dtype=torch.float64)
Loss:
tensor(2.0850, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
54
Accuracy:
tensor(0.1917, dtype=torch.float64)
Loss:
tensor(2.1458, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
55
Accuracy:
tensor(0.2208, dtype=torch.float64)
Loss:
tensor(2.0861, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
56
Accuracy:
tensor(0.2333, dtype=torch.float64)
Loss:
tensor(2.0917, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
57
Accuracy:
tensor(0.2417, dtype=torch.float64)
Loss:
tensor(2.0908, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
58
Accuracy:
tensor(0.2083, dtype=torch.float64)
Loss:
tensor(2.2014, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
59
Accuracy:
tensor(0.1833, dtype=torch.float64)
Loss:
tensor(2.1424, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
60
########################
Validation Accuracy:
tensor(0.2133, dtype=torch.float64)
########################
Accuracy:
tensor(0.2333, dtype=torch.float64)
Loss:
tensor(2.0685, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
61
Accuracy:
tensor(0.2208, dtype=torch.float64)
Loss:
tensor(2.1132, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
62
Accuracy:
tensor(0.1667, dtype=torch.float64)
Loss:
tensor(2.1374, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
63
Accuracy:
tensor(0.1750, dtype=torch.float64)
Loss:
tensor(2.1637, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
64
Accuracy:
tensor(0.2292, dtype=torch.float64)
Loss:
tensor(2.0733, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
65
Accuracy:
tensor(0.1833, dtype=torch.float64)
Loss:
tensor(2.1108, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
66
Accuracy:
tensor(0.2250, dtype=torch.float64)
Loss:
tensor(2.1150, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
67
Accuracy:
tensor(0.2292, dtype=torch.float64)
Loss:
tensor(2.0485, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
68
Accuracy:
tensor(0.2417, dtype=torch.float64)
Loss:
tensor(2.1370, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
69
Accuracy:
tensor(0.2333, dtype=torch.float64)
Loss:
tensor(2.0957, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
70
Accuracy:
tensor(0.1542, dtype=torch.float64)
Loss:
tensor(2.1910, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
71
Accuracy:
tensor(0.2292, dtype=torch.float64)
Loss:
tensor(2.1250, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
72
Accuracy:
tensor(0.2125, dtype=torch.float64)
Loss:
tensor(2.1248, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
73
Accuracy:
tensor(0.2542, dtype=torch.float64)
Loss:
tensor(2.0257, grad_fn=<NllLossBackward>)
Epoch:
0
Batch:
74
Accuracy:
tensor(0.3000, dtype=torch.float64)
Loss:
tensor(1.9978, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
0
########################
Validation Accuracy:
tensor(0.2217, dtype=torch.float64)
########################
Accuracy:
tensor(0.2708, dtype=torch.float64)
Loss:
tensor(2.0417, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
1
Accuracy:
tensor(0.3167, dtype=torch.float64)
Loss:
tensor(1.9986, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
2
Accuracy:
tensor(0.2500, dtype=torch.float64)
Loss:
tensor(1.9850, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
3
Accuracy:
tensor(0.2208, dtype=torch.float64)
Loss:
tensor(2.0333, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
4
Accuracy:
tensor(0.3042, dtype=torch.float64)
Loss:
tensor(1.9517, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
5
Accuracy:
tensor(0.3000, dtype=torch.float64)
Loss:
tensor(2.0286, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
6
Accuracy:
tensor(0.2542, dtype=torch.float64)
Loss:
tensor(2.0185, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
7
Accuracy:
tensor(0.3542, dtype=torch.float64)
Loss:
tensor(1.9182, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
8
Accuracy:
tensor(0.2500, dtype=torch.float64)
Loss:
tensor(2.0043, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
9
Accuracy:
tensor(0.2125, dtype=torch.float64)
Loss:
tensor(1.9722, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
10
Accuracy:
tensor(0.3208, dtype=torch.float64)
Loss:
tensor(1.9904, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
11
Accuracy:
tensor(0.2792, dtype=torch.float64)
Loss:
tensor(2.0617, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
12
Accuracy:
tensor(0.3542, dtype=torch.float64)
Loss:
tensor(1.9647, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
13
Accuracy:
tensor(0.3458, dtype=torch.float64)
Loss:
tensor(1.9563, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
14
Accuracy:
tensor(0.2167, dtype=torch.float64)
Loss:
tensor(2.0889, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
15
Accuracy:
tensor(0.2458, dtype=torch.float64)
Loss:
tensor(1.9933, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
16
Accuracy:
tensor(0.3167, dtype=torch.float64)
Loss:
tensor(2.0291, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
17
Accuracy:
tensor(0.2750, dtype=torch.float64)
Loss:
tensor(2.0512, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
18
Accuracy:
tensor(0.3250, dtype=torch.float64)
Loss:
tensor(1.9801, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
19
Accuracy:
tensor(0.3000, dtype=torch.float64)
Loss:
tensor(1.9443, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
20
########################
Validation Accuracy:
tensor(0.2317, dtype=torch.float64)
########################
Accuracy:
tensor(0.2833, dtype=torch.float64)
Loss:
tensor(1.9311, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
21
Accuracy:
tensor(0.2292, dtype=torch.float64)
Loss:
tensor(1.9948, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
22
Accuracy:
tensor(0.2833, dtype=torch.float64)
Loss:
tensor(1.9451, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
23
Accuracy:
tensor(0.3708, dtype=torch.float64)
Loss:
tensor(1.9361, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
24
Accuracy:
tensor(0.3458, dtype=torch.float64)
Loss:
tensor(1.9245, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
25
Accuracy:
tensor(0.3375, dtype=torch.float64)
Loss:
tensor(2.0149, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
26
Accuracy:
tensor(0.3000, dtype=torch.float64)
Loss:
tensor(1.9537, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
27
Accuracy:
tensor(0.2708, dtype=torch.float64)
Loss:
tensor(1.9935, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
28
Accuracy:
tensor(0.2792, dtype=torch.float64)
Loss:
tensor(1.9754, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
29
Accuracy:
tensor(0.2958, dtype=torch.float64)
Loss:
tensor(2.0006, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
30
Accuracy:
tensor(0.3500, dtype=torch.float64)
Loss:
tensor(1.9686, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
31
Accuracy:
tensor(0.2958, dtype=torch.float64)
Loss:
tensor(1.9588, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
32
Accuracy:
tensor(0.2833, dtype=torch.float64)
Loss:
tensor(2.0093, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
33
Accuracy:
tensor(0.2000, dtype=torch.float64)
Loss:
tensor(2.0942, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
34
Accuracy:
tensor(0.3208, dtype=torch.float64)
Loss:
tensor(2.0009, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
35
Accuracy:
tensor(0.3042, dtype=torch.float64)
Loss:
tensor(2.0005, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
36
Accuracy:
tensor(0.2375, dtype=torch.float64)
Loss:
tensor(2.0783, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
37
Accuracy:
tensor(0.3083, dtype=torch.float64)
Loss:
tensor(1.9166, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
38
Accuracy:
tensor(0.2167, dtype=torch.float64)
Loss:
tensor(2.0633, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
39
Accuracy:
tensor(0.3042, dtype=torch.float64)
Loss:
tensor(1.9290, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
40
########################
Validation Accuracy:
tensor(0.2567, dtype=torch.float64)
########################
Accuracy:
tensor(0.3500, dtype=torch.float64)
Loss:
tensor(1.9409, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
41
Accuracy:
tensor(0.3375, dtype=torch.float64)
Loss:
tensor(1.9436, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
42
Accuracy:
tensor(0.2708, dtype=torch.float64)
Loss:
tensor(1.9602, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
43
Accuracy:
tensor(0.2917, dtype=torch.float64)
Loss:
tensor(1.9853, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
44
Accuracy:
tensor(0.3000, dtype=torch.float64)
Loss:
tensor(1.9376, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
45
Accuracy:
tensor(0.3208, dtype=torch.float64)
Loss:
tensor(1.9140, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
46
Accuracy:
tensor(0.2917, dtype=torch.float64)
Loss:
tensor(1.9812, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
47
Accuracy:
tensor(0.2833, dtype=torch.float64)
Loss:
tensor(1.9768, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
48
Accuracy:
tensor(0.3125, dtype=torch.float64)
Loss:
tensor(1.9739, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
49
Accuracy:
tensor(0.3167, dtype=torch.float64)
Loss:
tensor(1.9348, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
50
Accuracy:
tensor(0.3000, dtype=torch.float64)
Loss:
tensor(1.9345, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
51
Accuracy:
tensor(0.3250, dtype=torch.float64)
Loss:
tensor(1.8772, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
52
Accuracy:
tensor(0.3125, dtype=torch.float64)
Loss:
tensor(1.9707, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
53
Accuracy:
tensor(0.2833, dtype=torch.float64)
Loss:
tensor(1.9310, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
54
Accuracy:
tensor(0.3208, dtype=torch.float64)
Loss:
tensor(1.9658, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
55
Accuracy:
tensor(0.3625, dtype=torch.float64)
Loss:
tensor(1.9120, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
56
Accuracy:
tensor(0.1917, dtype=torch.float64)
Loss:
tensor(2.0464, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
57
Accuracy:
tensor(0.2667, dtype=torch.float64)
Loss:
tensor(1.9667, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
58
Accuracy:
tensor(0.3417, dtype=torch.float64)
Loss:
tensor(1.9170, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
59
Accuracy:
tensor(0.2333, dtype=torch.float64)
Loss:
tensor(2.0100, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
60
########################
Validation Accuracy:
tensor(0.2450, dtype=torch.float64)
########################
Accuracy:
tensor(0.3292, dtype=torch.float64)
Loss:
tensor(1.9590, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
61
Accuracy:
tensor(0.3250, dtype=torch.float64)
Loss:
tensor(1.9096, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
62
Accuracy:
tensor(0.2667, dtype=torch.float64)
Loss:
tensor(2.0287, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
63
Accuracy:
tensor(0.2458, dtype=torch.float64)
Loss:
tensor(1.9762, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
64
Accuracy:
tensor(0.3750, dtype=torch.float64)
Loss:
tensor(1.9022, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
65
Accuracy:
tensor(0.3083, dtype=torch.float64)
Loss:
tensor(1.9708, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
66
Accuracy:
tensor(0.3042, dtype=torch.float64)
Loss:
tensor(1.9357, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
67
Accuracy:
tensor(0.2958, dtype=torch.float64)
Loss:
tensor(1.8672, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
68
Accuracy:
tensor(0.3167, dtype=torch.float64)
Loss:
tensor(1.8718, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
69
Accuracy:
tensor(0.2792, dtype=torch.float64)
Loss:
tensor(1.9545, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
70
Accuracy:
tensor(0.3292, dtype=torch.float64)
Loss:
tensor(1.9337, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
71
Accuracy:
tensor(0.2417, dtype=torch.float64)
Loss:
tensor(1.9720, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
72
Accuracy:
tensor(0.3083, dtype=torch.float64)
Loss:
tensor(1.9196, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
73
Accuracy:
tensor(0.3250, dtype=torch.float64)
Loss:
tensor(1.9202, grad_fn=<NllLossBackward>)
Epoch:
1
Batch:
74
Accuracy:
tensor(0.4583, dtype=torch.float64)
Loss:
tensor(1.7485, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
0
########################
Validation Accuracy:
tensor(0.2700, dtype=torch.float64)
########################
Accuracy:
tensor(0.3583, dtype=torch.float64)
Loss:
tensor(1.8449, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
1
Accuracy:
tensor(0.4208, dtype=torch.float64)
Loss:
tensor(1.7906, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
2
Accuracy:
tensor(0.4083, dtype=torch.float64)
Loss:
tensor(1.8773, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
3
Accuracy:
tensor(0.4167, dtype=torch.float64)
Loss:
tensor(1.7926, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
4
Accuracy:
tensor(0.4458, dtype=torch.float64)
Loss:
tensor(1.8039, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
5
Accuracy:
tensor(0.4292, dtype=torch.float64)
Loss:
tensor(1.7646, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
6
Accuracy:
tensor(0.4458, dtype=torch.float64)
Loss:
tensor(1.7878, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
7
Accuracy:
tensor(0.4250, dtype=torch.float64)
Loss:
tensor(1.8076, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
8
Accuracy:
tensor(0.3250, dtype=torch.float64)
Loss:
tensor(1.7923, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
9
Accuracy:
tensor(0.4208, dtype=torch.float64)
Loss:
tensor(1.7374, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
10
Accuracy:
tensor(0.4083, dtype=torch.float64)
Loss:
tensor(1.7708, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
11
Accuracy:
tensor(0.3750, dtype=torch.float64)
Loss:
tensor(1.8050, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
12
Accuracy:
tensor(0.3250, dtype=torch.float64)
Loss:
tensor(1.8655, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
13
Accuracy:
tensor(0.3708, dtype=torch.float64)
Loss:
tensor(1.7936, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
14
Accuracy:
tensor(0.4333, dtype=torch.float64)
Loss:
tensor(1.8081, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
15
Accuracy:
tensor(0.2792, dtype=torch.float64)
Loss:
tensor(1.9702, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
16
Accuracy:
tensor(0.4250, dtype=torch.float64)
Loss:
tensor(1.7945, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
17
Accuracy:
tensor(0.3125, dtype=torch.float64)
Loss:
tensor(1.8750, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
18
Accuracy:
tensor(0.3667, dtype=torch.float64)
Loss:
tensor(1.7864, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
19
Accuracy:
tensor(0.3583, dtype=torch.float64)
Loss:
tensor(1.8088, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
20
########################
Validation Accuracy:
tensor(0.2500, dtype=torch.float64)
########################
Accuracy:
tensor(0.3958, dtype=torch.float64)
Loss:
tensor(1.7648, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
21
Accuracy:
tensor(0.4000, dtype=torch.float64)
Loss:
tensor(1.7188, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
22
Accuracy:
tensor(0.3583, dtype=torch.float64)
Loss:
tensor(1.8307, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
23
Accuracy:
tensor(0.4000, dtype=torch.float64)
Loss:
tensor(1.8494, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
24
Accuracy:
tensor(0.3000, dtype=torch.float64)
Loss:
tensor(1.8890, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
25
Accuracy:
tensor(0.3750, dtype=torch.float64)
Loss:
tensor(1.8042, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
26
Accuracy:
tensor(0.4083, dtype=torch.float64)
Loss:
tensor(1.7735, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
27
Accuracy:
tensor(0.3083, dtype=torch.float64)
Loss:
tensor(1.8606, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
28
Accuracy:
tensor(0.4042, dtype=torch.float64)
Loss:
tensor(1.7609, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
29
Accuracy:
tensor(0.3917, dtype=torch.float64)
Loss:
tensor(1.7761, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
30
Accuracy:
tensor(0.3458, dtype=torch.float64)
Loss:
tensor(1.8288, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
31
Accuracy:
tensor(0.3917, dtype=torch.float64)
Loss:
tensor(1.7791, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
32
Accuracy:
tensor(0.3208, dtype=torch.float64)
Loss:
tensor(1.8022, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
33
Accuracy:
tensor(0.3917, dtype=torch.float64)
Loss:
tensor(1.7429, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
34
Accuracy:
tensor(0.3833, dtype=torch.float64)
Loss:
tensor(1.8766, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
35
Accuracy:
tensor(0.3708, dtype=torch.float64)
Loss:
tensor(1.7788, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
36
Accuracy:
tensor(0.4375, dtype=torch.float64)
Loss:
tensor(1.7094, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
37
Accuracy:
tensor(0.3917, dtype=torch.float64)
Loss:
tensor(1.7829, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
38
Accuracy:
tensor(0.4292, dtype=torch.float64)
Loss:
tensor(1.7406, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
39
Accuracy:
tensor(0.3958, dtype=torch.float64)
Loss:
tensor(1.7732, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
40
########################
Validation Accuracy:
tensor(0.2600, dtype=torch.float64)
########################
Accuracy:
tensor(0.3958, dtype=torch.float64)
Loss:
tensor(1.7995, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
41
Accuracy:
tensor(0.4125, dtype=torch.float64)
Loss:
tensor(1.7178, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
42
Accuracy:
tensor(0.3708, dtype=torch.float64)
Loss:
tensor(1.7658, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
43
Accuracy:
tensor(0.3958, dtype=torch.float64)
Loss:
tensor(1.7172, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
44
Accuracy:
tensor(0.3333, dtype=torch.float64)
Loss:
tensor(1.8218, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
45
Accuracy:
tensor(0.3625, dtype=torch.float64)
Loss:
tensor(1.7171, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
46
Accuracy:
tensor(0.4125, dtype=torch.float64)
Loss:
tensor(1.7521, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
47
Accuracy:
tensor(0.3500, dtype=torch.float64)
Loss:
tensor(1.7939, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
48
Accuracy:
tensor(0.4042, dtype=torch.float64)
Loss:
tensor(1.7191, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
49
Accuracy:
tensor(0.4000, dtype=torch.float64)
Loss:
tensor(1.7556, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
50
Accuracy:
tensor(0.3500, dtype=torch.float64)
Loss:
tensor(1.7939, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
51
Accuracy:
tensor(0.3292, dtype=torch.float64)
Loss:
tensor(1.8367, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
52
Accuracy:
tensor(0.4000, dtype=torch.float64)
Loss:
tensor(1.7514, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
53
Accuracy:
tensor(0.3708, dtype=torch.float64)
Loss:
tensor(1.7698, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
54
Accuracy:
tensor(0.3750, dtype=torch.float64)
Loss:
tensor(1.8006, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
55
Accuracy:
tensor(0.3583, dtype=torch.float64)
Loss:
tensor(1.7769, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
56
Accuracy:
tensor(0.3583, dtype=torch.float64)
Loss:
tensor(1.7378, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
57
Accuracy:
tensor(0.4458, dtype=torch.float64)
Loss:
tensor(1.6306, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
58
Accuracy:
tensor(0.3792, dtype=torch.float64)
Loss:
tensor(1.7382, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
59
Accuracy:
tensor(0.3292, dtype=torch.float64)
Loss:
tensor(1.7841, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
60
########################
Validation Accuracy:
tensor(0.2783, dtype=torch.float64)
########################
Accuracy:
tensor(0.3542, dtype=torch.float64)
Loss:
tensor(1.7813, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
61
Accuracy:
tensor(0.3250, dtype=torch.float64)
Loss:
tensor(1.7462, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
62
Accuracy:
tensor(0.4500, dtype=torch.float64)
Loss:
tensor(1.7202, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
63
Accuracy:
tensor(0.2958, dtype=torch.float64)
Loss:
tensor(1.9448, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
64
Accuracy:
tensor(0.3917, dtype=torch.float64)
Loss:
tensor(1.7547, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
65
Accuracy:
tensor(0.3500, dtype=torch.float64)
Loss:
tensor(1.8875, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
66
Accuracy:
tensor(0.3917, dtype=torch.float64)
Loss:
tensor(1.7725, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
67
Accuracy:
tensor(0.2917, dtype=torch.float64)
Loss:
tensor(1.8666, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
68
Accuracy:
tensor(0.3958, dtype=torch.float64)
Loss:
tensor(1.6874, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
69
Accuracy:
tensor(0.3833, dtype=torch.float64)
Loss:
tensor(1.7837, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
70
Accuracy:
tensor(0.3458, dtype=torch.float64)
Loss:
tensor(1.7909, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
71
Accuracy:
tensor(0.4000, dtype=torch.float64)
Loss:
tensor(1.7299, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
72
Accuracy:
tensor(0.3500, dtype=torch.float64)
Loss:
tensor(1.7947, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
73
Accuracy:
tensor(0.3583, dtype=torch.float64)
Loss:
tensor(1.7650, grad_fn=<NllLossBackward>)
Epoch:
2
Batch:
74
Accuracy:
tensor(0.3583, dtype=torch.float64)
Loss:
tensor(1.7727, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
0
########################
Validation Accuracy:
tensor(0.2767, dtype=torch.float64)
########################
Accuracy:
tensor(0.5333, dtype=torch.float64)
Loss:
tensor(1.5706, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
1
Accuracy:
tensor(0.4625, dtype=torch.float64)
Loss:
tensor(1.6077, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
2
Accuracy:
tensor(0.4542, dtype=torch.float64)
Loss:
tensor(1.6627, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
3
Accuracy:
tensor(0.4708, dtype=torch.float64)
Loss:
tensor(1.6470, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
4
Accuracy:
tensor(0.5042, dtype=torch.float64)
Loss:
tensor(1.6056, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
5
Accuracy:
tensor(0.4833, dtype=torch.float64)
Loss:
tensor(1.5374, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
6
Accuracy:
tensor(0.3083, dtype=torch.float64)
Loss:
tensor(1.9241, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
7
Accuracy:
tensor(0.5042, dtype=torch.float64)
Loss:
tensor(1.6297, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
8
Accuracy:
tensor(0.4042, dtype=torch.float64)
Loss:
tensor(1.6757, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
9
Accuracy:
tensor(0.3375, dtype=torch.float64)
Loss:
tensor(1.8296, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
10
Accuracy:
tensor(0.4625, dtype=torch.float64)
Loss:
tensor(1.6365, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
11
Accuracy:
tensor(0.3292, dtype=torch.float64)
Loss:
tensor(1.7983, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
12
Accuracy:
tensor(0.4750, dtype=torch.float64)
Loss:
tensor(1.6132, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
13
Accuracy:
tensor(0.4167, dtype=torch.float64)
Loss:
tensor(1.5908, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
14
Accuracy:
tensor(0.4333, dtype=torch.float64)
Loss:
tensor(1.6336, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
15
Accuracy:
tensor(0.4542, dtype=torch.float64)
Loss:
tensor(1.5924, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
16
Accuracy:
tensor(0.4083, dtype=torch.float64)
Loss:
tensor(1.7147, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
17
Accuracy:
tensor(0.4500, dtype=torch.float64)
Loss:
tensor(1.7035, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
18
Accuracy:
tensor(0.5125, dtype=torch.float64)
Loss:
tensor(1.6087, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
19
Accuracy:
tensor(0.4542, dtype=torch.float64)
Loss:
tensor(1.6580, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
20
########################
Validation Accuracy:
tensor(0.2833, dtype=torch.float64)
########################
Accuracy:
tensor(0.4750, dtype=torch.float64)
Loss:
tensor(1.5581, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
21
Accuracy:
tensor(0.4375, dtype=torch.float64)
Loss:
tensor(1.6367, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
22
Accuracy:
tensor(0.3792, dtype=torch.float64)
Loss:
tensor(1.7324, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
23
Accuracy:
tensor(0.4250, dtype=torch.float64)
Loss:
tensor(1.6791, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
24
Accuracy:
tensor(0.4958, dtype=torch.float64)
Loss:
tensor(1.5959, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
25
Accuracy:
tensor(0.4250, dtype=torch.float64)
Loss:
tensor(1.6205, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
26
Accuracy:
tensor(0.4583, dtype=torch.float64)
Loss:
tensor(1.5569, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
27
Accuracy:
tensor(0.4750, dtype=torch.float64)
Loss:
tensor(1.5928, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
28
Accuracy:
tensor(0.4250, dtype=torch.float64)
Loss:
tensor(1.6406, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
29
Accuracy:
tensor(0.4917, dtype=torch.float64)
Loss:
tensor(1.6001, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
30
Accuracy:
tensor(0.4000, dtype=torch.float64)
Loss:
tensor(1.5996, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
31
Accuracy:
tensor(0.4583, dtype=torch.float64)
Loss:
tensor(1.5235, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
32
Accuracy:
tensor(0.3833, dtype=torch.float64)
Loss:
tensor(1.6633, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
33
Accuracy:
tensor(0.4750, dtype=torch.float64)
Loss:
tensor(1.6194, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
34
Accuracy:
tensor(0.4500, dtype=torch.float64)
Loss:
tensor(1.5653, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
35
Accuracy:
tensor(0.4542, dtype=torch.float64)
Loss:
tensor(1.6285, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
36
Accuracy:
tensor(0.4667, dtype=torch.float64)
Loss:
tensor(1.5709, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
37
Accuracy:
tensor(0.3708, dtype=torch.float64)
Loss:
tensor(1.7375, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
38
Accuracy:
tensor(0.4583, dtype=torch.float64)
Loss:
tensor(1.6050, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
39
Accuracy:
tensor(0.4208, dtype=torch.float64)
Loss:
tensor(1.6003, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
40
########################
Validation Accuracy:
tensor(0.2817, dtype=torch.float64)
########################
Accuracy:
tensor(0.4375, dtype=torch.float64)
Loss:
tensor(1.6021, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
41
Accuracy:
tensor(0.4792, dtype=torch.float64)
Loss:
tensor(1.5180, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
42
Accuracy:
tensor(0.3625, dtype=torch.float64)
Loss:
tensor(1.7357, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
43
Accuracy:
tensor(0.4375, dtype=torch.float64)
Loss:
tensor(1.6135, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
44
Accuracy:
tensor(0.4333, dtype=torch.float64)
Loss:
tensor(1.6657, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
45
Accuracy:
tensor(0.4208, dtype=torch.float64)
Loss:
tensor(1.6431, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
46
Accuracy:
tensor(0.4625, dtype=torch.float64)
Loss:
tensor(1.5455, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
47
Accuracy:
tensor(0.4375, dtype=torch.float64)
Loss:
tensor(1.5842, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
48
Accuracy:
tensor(0.4375, dtype=torch.float64)
Loss:
tensor(1.6033, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
49
Accuracy:
tensor(0.4500, dtype=torch.float64)
Loss:
tensor(1.6073, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
50
Accuracy:
tensor(0.4375, dtype=torch.float64)
Loss:
tensor(1.6136, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
51
Accuracy:
tensor(0.5083, dtype=torch.float64)
Loss:
tensor(1.5227, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
52
Accuracy:
tensor(0.3625, dtype=torch.float64)
Loss:
tensor(1.6860, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
53
Accuracy:
tensor(0.3875, dtype=torch.float64)
Loss:
tensor(1.6991, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
54
Accuracy:
tensor(0.4417, dtype=torch.float64)
Loss:
tensor(1.6181, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
55
Accuracy:
tensor(0.4042, dtype=torch.float64)
Loss:
tensor(1.6541, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
56
Accuracy:
tensor(0.4250, dtype=torch.float64)
Loss:
tensor(1.6582, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
57
Accuracy:
tensor(0.4542, dtype=torch.float64)
Loss:
tensor(1.6567, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
58
Accuracy:
tensor(0.3667, dtype=torch.float64)
Loss:
tensor(1.7075, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
59
Accuracy:
tensor(0.4458, dtype=torch.float64)
Loss:
tensor(1.6116, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
60
########################
Validation Accuracy:
tensor(0.2967, dtype=torch.float64)
########################
Accuracy:
tensor(0.4583, dtype=torch.float64)
Loss:
tensor(1.6069, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
61
Accuracy:
tensor(0.4333, dtype=torch.float64)
Loss:
tensor(1.5828, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
62
Accuracy:
tensor(0.4542, dtype=torch.float64)
Loss:
tensor(1.5791, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
63
Accuracy:
tensor(0.3917, dtype=torch.float64)
Loss:
tensor(1.6350, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
64
Accuracy:
tensor(0.4000, dtype=torch.float64)
Loss:
tensor(1.6642, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
65
Accuracy:
tensor(0.4250, dtype=torch.float64)
Loss:
tensor(1.6416, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
66
Accuracy:
tensor(0.3917, dtype=torch.float64)
Loss:
tensor(1.7476, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
67
Accuracy:
tensor(0.4458, dtype=torch.float64)
Loss:
tensor(1.5603, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
68
Accuracy:
tensor(0.4333, dtype=torch.float64)
Loss:
tensor(1.5704, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
69
Accuracy:
tensor(0.4458, dtype=torch.float64)
Loss:
tensor(1.5600, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
70
Accuracy:
tensor(0.4833, dtype=torch.float64)
Loss:
tensor(1.4543, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
71
Accuracy:
tensor(0.4500, dtype=torch.float64)
Loss:
tensor(1.5255, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
72
Accuracy:
tensor(0.4708, dtype=torch.float64)
Loss:
tensor(1.5088, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
73
Accuracy:
tensor(0.4458, dtype=torch.float64)
Loss:
tensor(1.5901, grad_fn=<NllLossBackward>)
Epoch:
3
Batch:
74
Accuracy:
tensor(0.5292, dtype=torch.float64)
Loss:
tensor(1.4482, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
0
########################
Validation Accuracy:
tensor(0.2933, dtype=torch.float64)
########################
Accuracy:
tensor(0.5417, dtype=torch.float64)
Loss:
tensor(1.4351, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
1
Accuracy:
tensor(0.5708, dtype=torch.float64)
Loss:
tensor(1.3942, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
2
Accuracy:
tensor(0.5500, dtype=torch.float64)
Loss:
tensor(1.4766, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
3
Accuracy:
tensor(0.5042, dtype=torch.float64)
Loss:
tensor(1.5133, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
4
Accuracy:
tensor(0.5042, dtype=torch.float64)
Loss:
tensor(1.4851, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
5
Accuracy:
tensor(0.5250, dtype=torch.float64)
Loss:
tensor(1.4336, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
6
Accuracy:
tensor(0.5542, dtype=torch.float64)
Loss:
tensor(1.4504, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
7
Accuracy:
tensor(0.5750, dtype=torch.float64)
Loss:
tensor(1.3311, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
8
Accuracy:
tensor(0.5125, dtype=torch.float64)
Loss:
tensor(1.4411, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
9
Accuracy:
tensor(0.6083, dtype=torch.float64)
Loss:
tensor(1.2910, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
10
Accuracy:
tensor(0.4667, dtype=torch.float64)
Loss:
tensor(1.4671, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
11
Accuracy:
tensor(0.4833, dtype=torch.float64)
Loss:
tensor(1.4903, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
12
Accuracy:
tensor(0.4792, dtype=torch.float64)
Loss:
tensor(1.5046, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
13
Accuracy:
tensor(0.5708, dtype=torch.float64)
Loss:
tensor(1.3999, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
14
Accuracy:
tensor(0.5333, dtype=torch.float64)
Loss:
tensor(1.3937, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
15
Accuracy:
tensor(0.4792, dtype=torch.float64)
Loss:
tensor(1.4654, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
16
Accuracy:
tensor(0.5250, dtype=torch.float64)
Loss:
tensor(1.3773, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
17
Accuracy:
tensor(0.5208, dtype=torch.float64)
Loss:
tensor(1.4448, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
18
Accuracy:
tensor(0.5958, dtype=torch.float64)
Loss:
tensor(1.3926, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
19
Accuracy:
tensor(0.5125, dtype=torch.float64)
Loss:
tensor(1.4770, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
20
########################
Validation Accuracy:
tensor(0.3050, dtype=torch.float64)
########################
Accuracy:
tensor(0.5292, dtype=torch.float64)
Loss:
tensor(1.4610, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
21
Accuracy:
tensor(0.5292, dtype=torch.float64)
Loss:
tensor(1.3751, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
22
Accuracy:
tensor(0.5125, dtype=torch.float64)
Loss:
tensor(1.4630, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
23
Accuracy:
tensor(0.4333, dtype=torch.float64)
Loss:
tensor(1.5065, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
24
Accuracy:
tensor(0.5250, dtype=torch.float64)
Loss:
tensor(1.4189, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
25
Accuracy:
tensor(0.4958, dtype=torch.float64)
Loss:
tensor(1.4911, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
26
Accuracy:
tensor(0.4833, dtype=torch.float64)
Loss:
tensor(1.5624, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
27
Accuracy:
tensor(0.5333, dtype=torch.float64)
Loss:
tensor(1.4364, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
28
Accuracy:
tensor(0.4458, dtype=torch.float64)
Loss:
tensor(1.5698, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
29
Accuracy:
tensor(0.4792, dtype=torch.float64)
Loss:
tensor(1.4355, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
30
Accuracy:
tensor(0.5250, dtype=torch.float64)
Loss:
tensor(1.4422, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
31
Accuracy:
tensor(0.4292, dtype=torch.float64)
Loss:
tensor(1.5908, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
32
Accuracy:
tensor(0.4583, dtype=torch.float64)
Loss:
tensor(1.5051, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
33
Accuracy:
tensor(0.4292, dtype=torch.float64)
Loss:
tensor(1.4963, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
34
Accuracy:
tensor(0.5042, dtype=torch.float64)
Loss:
tensor(1.4934, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
35
Accuracy:
tensor(0.4875, dtype=torch.float64)
Loss:
tensor(1.5076, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
36
Accuracy:
tensor(0.5125, dtype=torch.float64)
Loss:
tensor(1.5464, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
37
Accuracy:
tensor(0.4750, dtype=torch.float64)
Loss:
tensor(1.5050, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
38
Accuracy:
tensor(0.4375, dtype=torch.float64)
Loss:
tensor(1.5433, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
39
Accuracy:
tensor(0.3667, dtype=torch.float64)
Loss:
tensor(1.5849, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
40
########################
Validation Accuracy:
tensor(0.2950, dtype=torch.float64)
########################
Accuracy:
tensor(0.5333, dtype=torch.float64)
Loss:
tensor(1.3990, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
41
Accuracy:
tensor(0.4708, dtype=torch.float64)
Loss:
tensor(1.4867, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
42
Accuracy:
tensor(0.5458, dtype=torch.float64)
Loss:
tensor(1.4315, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
43
Accuracy:
tensor(0.4667, dtype=torch.float64)
Loss:
tensor(1.4933, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
44
Accuracy:
tensor(0.4583, dtype=torch.float64)
Loss:
tensor(1.5417, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
45
Accuracy:
tensor(0.5417, dtype=torch.float64)
Loss:
tensor(1.4232, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
46
Accuracy:
tensor(0.5042, dtype=torch.float64)
Loss:
tensor(1.4332, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
47
Accuracy:
tensor(0.4208, dtype=torch.float64)
Loss:
tensor(1.6155, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
48
Accuracy:
tensor(0.4958, dtype=torch.float64)
Loss:
tensor(1.4129, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
49
Accuracy:
tensor(0.4458, dtype=torch.float64)
Loss:
tensor(1.5357, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
50
Accuracy:
tensor(0.5250, dtype=torch.float64)
Loss:
tensor(1.4330, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
51
Accuracy:
tensor(0.5125, dtype=torch.float64)
Loss:
tensor(1.4957, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
52
Accuracy:
tensor(0.4833, dtype=torch.float64)
Loss:
tensor(1.4911, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
53
Accuracy:
tensor(0.4417, dtype=torch.float64)
Loss:
tensor(1.5498, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
54
Accuracy:
tensor(0.5917, dtype=torch.float64)
Loss:
tensor(1.3303, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
55
Accuracy:
tensor(0.5125, dtype=torch.float64)
Loss:
tensor(1.4205, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
56
Accuracy:
tensor(0.5000, dtype=torch.float64)
Loss:
tensor(1.4744, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
57
Accuracy:
tensor(0.4833, dtype=torch.float64)
Loss:
tensor(1.5064, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
58
Accuracy:
tensor(0.5083, dtype=torch.float64)
Loss:
tensor(1.4697, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
59
Accuracy:
tensor(0.4792, dtype=torch.float64)
Loss:
tensor(1.4862, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
60
########################
Validation Accuracy:
tensor(0.2883, dtype=torch.float64)
########################
Accuracy:
tensor(0.4667, dtype=torch.float64)
Loss:
tensor(1.5109, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
61
Accuracy:
tensor(0.4417, dtype=torch.float64)
Loss:
tensor(1.4861, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
62
Accuracy:
tensor(0.5375, dtype=torch.float64)
Loss:
tensor(1.4062, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
63
Accuracy:
tensor(0.5250, dtype=torch.float64)
Loss:
tensor(1.4533, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
64
Accuracy:
tensor(0.4917, dtype=torch.float64)
Loss:
tensor(1.4714, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
65
Accuracy:
tensor(0.5375, dtype=torch.float64)
Loss:
tensor(1.4255, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
66
Accuracy:
tensor(0.5167, dtype=torch.float64)
Loss:
tensor(1.3819, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
67
Accuracy:
tensor(0.5083, dtype=torch.float64)
Loss:
tensor(1.4646, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
68
Accuracy:
tensor(0.4750, dtype=torch.float64)
Loss:
tensor(1.4990, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
69
Accuracy:
tensor(0.4833, dtype=torch.float64)
Loss:
tensor(1.4565, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
70
Accuracy:
tensor(0.5208, dtype=torch.float64)
Loss:
tensor(1.3924, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
71
Accuracy:
tensor(0.5208, dtype=torch.float64)
Loss:
tensor(1.4143, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
72
Accuracy:
tensor(0.4625, dtype=torch.float64)
Loss:
tensor(1.5426, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
73
Accuracy:
tensor(0.5167, dtype=torch.float64)
Loss:
tensor(1.4336, grad_fn=<NllLossBackward>)
Epoch:
4
Batch:
74
Accuracy:
tensor(0.6333, dtype=torch.float64)
Loss:
tensor(1.2044, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
0
########################
Validation Accuracy:
tensor(0.2800, dtype=torch.float64)
########################
Accuracy:
tensor(0.6667, dtype=torch.float64)
Loss:
tensor(1.1815, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
1
Accuracy:
tensor(0.6625, dtype=torch.float64)
Loss:
tensor(1.2011, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
2
Accuracy:
tensor(0.4708, dtype=torch.float64)
Loss:
tensor(1.4478, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
3
Accuracy:
tensor(0.5917, dtype=torch.float64)
Loss:
tensor(1.2464, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
4
Accuracy:
tensor(0.5750, dtype=torch.float64)
Loss:
tensor(1.2892, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
5
Accuracy:
tensor(0.6083, dtype=torch.float64)
Loss:
tensor(1.2650, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
6
Accuracy:
tensor(0.4125, dtype=torch.float64)
Loss:
tensor(1.5788, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
7
Accuracy:
tensor(0.6167, dtype=torch.float64)
Loss:
tensor(1.2912, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
8
Accuracy:
tensor(0.5375, dtype=torch.float64)
Loss:
tensor(1.2969, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
9
Accuracy:
tensor(0.5208, dtype=torch.float64)
Loss:
tensor(1.4273, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
10
Accuracy:
tensor(0.6042, dtype=torch.float64)
Loss:
tensor(1.2928, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
11
Accuracy:
tensor(0.6000, dtype=torch.float64)
Loss:
tensor(1.2700, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
12
Accuracy:
tensor(0.5250, dtype=torch.float64)
Loss:
tensor(1.3402, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
13
Accuracy:
tensor(0.6167, dtype=torch.float64)
Loss:
tensor(1.2843, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
14
Accuracy:
tensor(0.6375, dtype=torch.float64)
Loss:
tensor(1.2548, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
15
Accuracy:
tensor(0.5875, dtype=torch.float64)
Loss:
tensor(1.3072, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
16
Accuracy:
tensor(0.6333, dtype=torch.float64)
Loss:
tensor(1.2207, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
17
Accuracy:
tensor(0.6125, dtype=torch.float64)
Loss:
tensor(1.2773, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
18
Accuracy:
tensor(0.5875, dtype=torch.float64)
Loss:
tensor(1.3346, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
19
Accuracy:
tensor(0.5917, dtype=torch.float64)
Loss:
tensor(1.3700, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
20
########################
Validation Accuracy:
tensor(0.2800, dtype=torch.float64)
########################
Accuracy:
tensor(0.5375, dtype=torch.float64)
Loss:
tensor(1.3954, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
21
Accuracy:
tensor(0.6417, dtype=torch.float64)
Loss:
tensor(1.2022, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
22
Accuracy:
tensor(0.6458, dtype=torch.float64)
Loss:
tensor(1.2640, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
23
Accuracy:
tensor(0.5333, dtype=torch.float64)
Loss:
tensor(1.3867, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
24
Accuracy:
tensor(0.5917, dtype=torch.float64)
Loss:
tensor(1.3042, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
25
Accuracy:
tensor(0.5750, dtype=torch.float64)
Loss:
tensor(1.2589, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
26
Accuracy:
tensor(0.6333, dtype=torch.float64)
Loss:
tensor(1.2055, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
27
Accuracy:
tensor(0.5500, dtype=torch.float64)
Loss:
tensor(1.3025, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
28
Accuracy:
tensor(0.5958, dtype=torch.float64)
Loss:
tensor(1.2925, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
29
Accuracy:
tensor(0.5417, dtype=torch.float64)
Loss:
tensor(1.3175, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
30
Accuracy:
tensor(0.6250, dtype=torch.float64)
Loss:
tensor(1.2343, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
31
Accuracy:
tensor(0.6000, dtype=torch.float64)
Loss:
tensor(1.2716, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
32
Accuracy:
tensor(0.5333, dtype=torch.float64)
Loss:
tensor(1.3450, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
33
Accuracy:
tensor(0.5833, dtype=torch.float64)
Loss:
tensor(1.2344, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
34
Accuracy:
tensor(0.5250, dtype=torch.float64)
Loss:
tensor(1.3572, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
35
Accuracy:
tensor(0.6000, dtype=torch.float64)
Loss:
tensor(1.2379, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
36
Accuracy:
tensor(0.5708, dtype=torch.float64)
Loss:
tensor(1.2834, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
37
Accuracy:
tensor(0.5750, dtype=torch.float64)
Loss:
tensor(1.2516, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
38
Accuracy:
tensor(0.5208, dtype=torch.float64)
Loss:
tensor(1.4073, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
39
Accuracy:
tensor(0.4917, dtype=torch.float64)
Loss:
tensor(1.3289, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
40
########################
Validation Accuracy:
tensor(0.2817, dtype=torch.float64)
########################
Accuracy:
tensor(0.5167, dtype=torch.float64)
Loss:
tensor(1.4268, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
41
Accuracy:
tensor(0.5792, dtype=torch.float64)
Loss:
tensor(1.2780, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
42
Accuracy:
tensor(0.5292, dtype=torch.float64)
Loss:
tensor(1.4193, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
43
Accuracy:
tensor(0.5750, dtype=torch.float64)
Loss:
tensor(1.3136, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
44
Accuracy:
tensor(0.5042, dtype=torch.float64)
Loss:
tensor(1.3835, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
45
Accuracy:
tensor(0.5333, dtype=torch.float64)
Loss:
tensor(1.2976, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
46
Accuracy:
tensor(0.5583, dtype=torch.float64)
Loss:
tensor(1.3389, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
47
Accuracy:
tensor(0.6000, dtype=torch.float64)
Loss:
tensor(1.2520, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
48
Accuracy:
tensor(0.4375, dtype=torch.float64)
Loss:
tensor(1.4476, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
49
Accuracy:
tensor(0.5375, dtype=torch.float64)
Loss:
tensor(1.3349, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
50
Accuracy:
tensor(0.4542, dtype=torch.float64)
Loss:
tensor(1.5132, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
51
Accuracy:
tensor(0.5333, dtype=torch.float64)
Loss:
tensor(1.3523, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
52
Accuracy:
tensor(0.6125, dtype=torch.float64)
Loss:
tensor(1.2675, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
53
Accuracy:
tensor(0.5583, dtype=torch.float64)
Loss:
tensor(1.3114, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
54
Accuracy:
tensor(0.5167, dtype=torch.float64)
Loss:
tensor(1.3835, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
55
Accuracy:
tensor(0.5375, dtype=torch.float64)
Loss:
tensor(1.3793, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
56
Accuracy:
tensor(0.4917, dtype=torch.float64)
Loss:
tensor(1.4656, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
57
Accuracy:
tensor(0.6250, dtype=torch.float64)
Loss:
tensor(1.2861, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
58
Accuracy:
tensor(0.5708, dtype=torch.float64)
Loss:
tensor(1.2619, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
59
Accuracy:
tensor(0.4792, dtype=torch.float64)
Loss:
tensor(1.4211, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
60
########################
Validation Accuracy:
tensor(0.2933, dtype=torch.float64)
########################
Accuracy:
tensor(0.5250, dtype=torch.float64)
Loss:
tensor(1.3878, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
61
Accuracy:
tensor(0.5875, dtype=torch.float64)
Loss:
tensor(1.2625, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
62
Accuracy:
tensor(0.4958, dtype=torch.float64)
Loss:
tensor(1.3929, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
63
Accuracy:
tensor(0.5333, dtype=torch.float64)
Loss:
tensor(1.3897, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
64
Accuracy:
tensor(0.5625, dtype=torch.float64)
Loss:
tensor(1.3088, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
65
Accuracy:
tensor(0.5292, dtype=torch.float64)
Loss:
tensor(1.3394, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
66
Accuracy:
tensor(0.5292, dtype=torch.float64)
Loss:
tensor(1.2988, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
67
Accuracy:
tensor(0.5417, dtype=torch.float64)
Loss:
tensor(1.3196, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
68
Accuracy:
tensor(0.5542, dtype=torch.float64)
Loss:
tensor(1.3452, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
69
Accuracy:
tensor(0.5500, dtype=torch.float64)
Loss:
tensor(1.2998, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
70
Accuracy:
tensor(0.5958, dtype=torch.float64)
Loss:
tensor(1.2433, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
71
Accuracy:
tensor(0.5750, dtype=torch.float64)
Loss:
tensor(1.2946, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
72
Accuracy:
tensor(0.6000, dtype=torch.float64)
Loss:
tensor(1.2018, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
73
Accuracy:
tensor(0.5625, dtype=torch.float64)
Loss:
tensor(1.3412, grad_fn=<NllLossBackward>)
Epoch:
5
Batch:
74
Accuracy:
tensor(0.6500, dtype=torch.float64)
Loss:
tensor(1.1364, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
0
########################
Validation Accuracy:
tensor(0.2817, dtype=torch.float64)
########################
Accuracy:
tensor(0.6667, dtype=torch.float64)
Loss:
tensor(1.1400, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
1
Accuracy:
tensor(0.6458, dtype=torch.float64)
Loss:
tensor(1.1770, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
2
Accuracy:
tensor(0.5625, dtype=torch.float64)
Loss:
tensor(1.2624, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
3
Accuracy:
tensor(0.6458, dtype=torch.float64)
Loss:
tensor(1.1639, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
4
Accuracy:
tensor(0.5750, dtype=torch.float64)
Loss:
tensor(1.2365, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
5
Accuracy:
tensor(0.6375, dtype=torch.float64)
Loss:
tensor(1.1423, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
6
Accuracy:
tensor(0.6125, dtype=torch.float64)
Loss:
tensor(1.1562, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
7
Accuracy:
tensor(0.6083, dtype=torch.float64)
Loss:
tensor(1.1267, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
8
Accuracy:
tensor(0.6792, dtype=torch.float64)
Loss:
tensor(1.0492, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
9
Accuracy:
tensor(0.5583, dtype=torch.float64)
Loss:
tensor(1.2360, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
10
Accuracy:
tensor(0.5708, dtype=torch.float64)
Loss:
tensor(1.2869, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
11
Accuracy:
tensor(0.6167, dtype=torch.float64)
Loss:
tensor(1.1679, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
12
Accuracy:
tensor(0.6333, dtype=torch.float64)
Loss:
tensor(1.1727, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
13
Accuracy:
tensor(0.6250, dtype=torch.float64)
Loss:
tensor(1.1931, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
14
Accuracy:
tensor(0.6250, dtype=torch.float64)
Loss:
tensor(1.2081, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
15
Accuracy:
tensor(0.6833, dtype=torch.float64)
Loss:
tensor(1.1602, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
16
Accuracy:
tensor(0.5792, dtype=torch.float64)
Loss:
tensor(1.2110, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
17
Accuracy:
tensor(0.5458, dtype=torch.float64)
Loss:
tensor(1.2154, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
18
Accuracy:
tensor(0.6292, dtype=torch.float64)
Loss:
tensor(1.1516, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
19
Accuracy:
tensor(0.6458, dtype=torch.float64)
Loss:
tensor(1.1709, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
20
########################
Validation Accuracy:
tensor(0.2667, dtype=torch.float64)
########################
Accuracy:
tensor(0.6250, dtype=torch.float64)
Loss:
tensor(1.1808, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
21
Accuracy:
tensor(0.6125, dtype=torch.float64)
Loss:
tensor(1.1331, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
22
Accuracy:
tensor(0.5708, dtype=torch.float64)
Loss:
tensor(1.2673, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
23
Accuracy:
tensor(0.6500, dtype=torch.float64)
Loss:
tensor(1.1178, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
24
Accuracy:
tensor(0.6417, dtype=torch.float64)
Loss:
tensor(1.1296, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
25
Accuracy:
tensor(0.6208, dtype=torch.float64)
Loss:
tensor(1.1664, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
26
Accuracy:
tensor(0.5583, dtype=torch.float64)
Loss:
tensor(1.2674, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
27
Accuracy:
tensor(0.6708, dtype=torch.float64)
Loss:
tensor(1.1376, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
28
Accuracy:
tensor(0.5417, dtype=torch.float64)
Loss:
tensor(1.2987, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
29
Accuracy:
tensor(0.5500, dtype=torch.float64)
Loss:
tensor(1.2642, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
30
Accuracy:
tensor(0.6083, dtype=torch.float64)
Loss:
tensor(1.2150, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
31
Accuracy:
tensor(0.5917, dtype=torch.float64)
Loss:
tensor(1.1924, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
32
Accuracy:
tensor(0.6167, dtype=torch.float64)
Loss:
tensor(1.2413, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
33
Accuracy:
tensor(0.6417, dtype=torch.float64)
Loss:
tensor(1.1896, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
34
Accuracy:
tensor(0.6667, dtype=torch.float64)
Loss:
tensor(1.0785, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
35
Accuracy:
tensor(0.5708, dtype=torch.float64)
Loss:
tensor(1.2986, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
36
Accuracy:
tensor(0.6333, dtype=torch.float64)
Loss:
tensor(1.1735, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
37
Accuracy:
tensor(0.5458, dtype=torch.float64)
Loss:
tensor(1.2525, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
38
Accuracy:
tensor(0.5875, dtype=torch.float64)
Loss:
tensor(1.1884, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
39
Accuracy:
tensor(0.5917, dtype=torch.float64)
Loss:
tensor(1.2611, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
40
########################
Validation Accuracy:
tensor(0.2850, dtype=torch.float64)
########################
Accuracy:
tensor(0.6125, dtype=torch.float64)
Loss:
tensor(1.1577, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
41
Accuracy:
tensor(0.6250, dtype=torch.float64)
Loss:
tensor(1.1702, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
42
Accuracy:
tensor(0.5917, dtype=torch.float64)
Loss:
tensor(1.1955, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
43
Accuracy:
tensor(0.6167, dtype=torch.float64)
Loss:
tensor(1.1953, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
44
Accuracy:
tensor(0.6625, dtype=torch.float64)
Loss:
tensor(1.0831, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
45
Accuracy:
tensor(0.6208, dtype=torch.float64)
Loss:
tensor(1.1572, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
46
Accuracy:
tensor(0.5875, dtype=torch.float64)
Loss:
tensor(1.2139, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
47
Accuracy:
tensor(0.6167, dtype=torch.float64)
Loss:
tensor(1.2236, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
48
Accuracy:
tensor(0.5375, dtype=torch.float64)
Loss:
tensor(1.2833, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
49
Accuracy:
tensor(0.5958, dtype=torch.float64)
Loss:
tensor(1.2668, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
50
Accuracy:
tensor(0.6542, dtype=torch.float64)
Loss:
tensor(1.1776, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
51
Accuracy:
tensor(0.6083, dtype=torch.float64)
Loss:
tensor(1.1780, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
52
Accuracy:
tensor(0.5792, dtype=torch.float64)
Loss:
tensor(1.2310, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
53
Accuracy:
tensor(0.5500, dtype=torch.float64)
Loss:
tensor(1.2624, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
54
Accuracy:
tensor(0.5875, dtype=torch.float64)
Loss:
tensor(1.1993, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
55
Accuracy:
tensor(0.5458, dtype=torch.float64)
Loss:
tensor(1.3213, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
56
Accuracy:
tensor(0.6042, dtype=torch.float64)
Loss:
tensor(1.1473, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
57
Accuracy:
tensor(0.6125, dtype=torch.float64)
Loss:
tensor(1.1694, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
58
Accuracy:
tensor(0.6250, dtype=torch.float64)
Loss:
tensor(1.1680, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
59
Accuracy:
tensor(0.6500, dtype=torch.float64)
Loss:
tensor(1.0686, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
60
########################
Validation Accuracy:
tensor(0.2717, dtype=torch.float64)
########################
Accuracy:
tensor(0.6042, dtype=torch.float64)
Loss:
tensor(1.1881, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
61
Accuracy:
tensor(0.6375, dtype=torch.float64)
Loss:
tensor(1.0882, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
62
Accuracy:
tensor(0.6167, dtype=torch.float64)
Loss:
tensor(1.1997, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
63
Accuracy:
tensor(0.5542, dtype=torch.float64)
Loss:
tensor(1.2461, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
64
Accuracy:
tensor(0.6125, dtype=torch.float64)
Loss:
tensor(1.1069, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
65
Accuracy:
tensor(0.5667, dtype=torch.float64)
Loss:
tensor(1.1948, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
66
Accuracy:
tensor(0.5583, dtype=torch.float64)
Loss:
tensor(1.2477, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
67
Accuracy:
tensor(0.6542, dtype=torch.float64)
Loss:
tensor(1.1461, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
68
Accuracy:
tensor(0.6125, dtype=torch.float64)
Loss:
tensor(1.0978, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
69
Accuracy:
tensor(0.5792, dtype=torch.float64)
Loss:
tensor(1.2685, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
70
Accuracy:
tensor(0.5708, dtype=torch.float64)
Loss:
tensor(1.2559, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
71
Accuracy:
tensor(0.6083, dtype=torch.float64)
Loss:
tensor(1.1554, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
72
Accuracy:
tensor(0.5792, dtype=torch.float64)
Loss:
tensor(1.2106, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
73
Accuracy:
tensor(0.6333, dtype=torch.float64)
Loss:
tensor(1.1428, grad_fn=<NllLossBackward>)
Epoch:
6
Batch:
74
Accuracy:
tensor(0.6250, dtype=torch.float64)
Loss:
tensor(1.0967, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
0
########################
Validation Accuracy:
tensor(0.2833, dtype=torch.float64)
########################
Accuracy:
tensor(0.6917, dtype=torch.float64)
Loss:
tensor(0.9697, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
1
Accuracy:
tensor(0.6917, dtype=torch.float64)
Loss:
tensor(1.0221, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
2
Accuracy:
tensor(0.6958, dtype=torch.float64)
Loss:
tensor(1.0225, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
3
Accuracy:
tensor(0.7167, dtype=torch.float64)
Loss:
tensor(0.9512, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
4
Accuracy:
tensor(0.7292, dtype=torch.float64)
Loss:
tensor(0.9927, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
5
Accuracy:
tensor(0.7208, dtype=torch.float64)
Loss:
tensor(0.9529, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
6
Accuracy:
tensor(0.7458, dtype=torch.float64)
Loss:
tensor(0.9542, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
7
Accuracy:
tensor(0.7042, dtype=torch.float64)
Loss:
tensor(0.9809, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
8
Accuracy:
tensor(0.6708, dtype=torch.float64)
Loss:
tensor(1.0346, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
9
Accuracy:
tensor(0.6917, dtype=torch.float64)
Loss:
tensor(1.0001, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
10
Accuracy:
tensor(0.7333, dtype=torch.float64)
Loss:
tensor(0.9575, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
11
Accuracy:
tensor(0.6958, dtype=torch.float64)
Loss:
tensor(1.0025, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
12
Accuracy:
tensor(0.6917, dtype=torch.float64)
Loss:
tensor(0.9484, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
13
Accuracy:
tensor(0.7417, dtype=torch.float64)
Loss:
tensor(0.9196, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
14
Accuracy:
tensor(0.6625, dtype=torch.float64)
Loss:
tensor(0.9909, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
15
Accuracy:
tensor(0.7083, dtype=torch.float64)
Loss:
tensor(0.9193, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
16
Accuracy:
tensor(0.6708, dtype=torch.float64)
Loss:
tensor(0.9943, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
17
Accuracy:
tensor(0.6583, dtype=torch.float64)
Loss:
tensor(1.1169, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
18
Accuracy:
tensor(0.7125, dtype=torch.float64)
Loss:
tensor(0.9430, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
19
Accuracy:
tensor(0.7125, dtype=torch.float64)
Loss:
tensor(0.9516, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
20
########################
Validation Accuracy:
tensor(0.3033, dtype=torch.float64)
########################
Accuracy:
tensor(0.7208, dtype=torch.float64)
Loss:
tensor(0.9123, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
21
Accuracy:
tensor(0.6708, dtype=torch.float64)
Loss:
tensor(1.0541, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
22
Accuracy:
tensor(0.6458, dtype=torch.float64)
Loss:
tensor(1.0299, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
23
Accuracy:
tensor(0.6375, dtype=torch.float64)
Loss:
tensor(1.1407, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
24
Accuracy:
tensor(0.7250, dtype=torch.float64)
Loss:
tensor(0.9054, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
25
Accuracy:
tensor(0.6917, dtype=torch.float64)
Loss:
tensor(0.9998, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
26
Accuracy:
tensor(0.7125, dtype=torch.float64)
Loss:
tensor(1.0135, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
27
Accuracy:
tensor(0.5792, dtype=torch.float64)
Loss:
tensor(1.1700, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
28
Accuracy:
tensor(0.6667, dtype=torch.float64)
Loss:
tensor(1.0885, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
29
Accuracy:
tensor(0.6167, dtype=torch.float64)
Loss:
tensor(1.0926, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
30
Accuracy:
tensor(0.6292, dtype=torch.float64)
Loss:
tensor(1.0800, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
31
Accuracy:
tensor(0.6792, dtype=torch.float64)
Loss:
tensor(0.9687, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
32
Accuracy:
tensor(0.5792, dtype=torch.float64)
Loss:
tensor(1.2287, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
33
Accuracy:
tensor(0.6625, dtype=torch.float64)
Loss:
tensor(1.0610, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
34
Accuracy:
tensor(0.6958, dtype=torch.float64)
Loss:
tensor(0.9958, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
35
Accuracy:
tensor(0.6375, dtype=torch.float64)
Loss:
tensor(1.1707, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
36
Accuracy:
tensor(0.5500, dtype=torch.float64)
Loss:
tensor(1.2776, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
37
Accuracy:
tensor(0.6250, dtype=torch.float64)
Loss:
tensor(1.0941, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
38
Accuracy:
tensor(0.5792, dtype=torch.float64)
Loss:
tensor(1.1213, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
39
Accuracy:
tensor(0.6333, dtype=torch.float64)
Loss:
tensor(1.1173, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
40
########################
Validation Accuracy:
tensor(0.2733, dtype=torch.float64)
########################
Accuracy:
tensor(0.6417, dtype=torch.float64)
Loss:
tensor(1.1267, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
41
Accuracy:
tensor(0.6250, dtype=torch.float64)
Loss:
tensor(1.1496, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
42
Accuracy:
tensor(0.6750, dtype=torch.float64)
Loss:
tensor(1.0255, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
43
Accuracy:
tensor(0.6583, dtype=torch.float64)
Loss:
tensor(1.0460, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
44
Accuracy:
tensor(0.5875, dtype=torch.float64)
Loss:
tensor(1.1385, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
45
Accuracy:
tensor(0.6083, dtype=torch.float64)
Loss:
tensor(1.1764, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
46
Accuracy:
tensor(0.6250, dtype=torch.float64)
Loss:
tensor(1.0944, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
47
Accuracy:
tensor(0.6958, dtype=torch.float64)
Loss:
tensor(0.9827, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
48
Accuracy:
tensor(0.6625, dtype=torch.float64)
Loss:
tensor(1.0420, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
49
Accuracy:
tensor(0.6458, dtype=torch.float64)
Loss:
tensor(1.0958, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
50
Accuracy:
tensor(0.6083, dtype=torch.float64)
Loss:
tensor(1.1346, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
51
Accuracy:
tensor(0.7000, dtype=torch.float64)
Loss:
tensor(1.0386, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
52
Accuracy:
tensor(0.5917, dtype=torch.float64)
Loss:
tensor(1.1733, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
53
Accuracy:
tensor(0.6583, dtype=torch.float64)
Loss:
tensor(1.0789, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
54
Accuracy:
tensor(0.6625, dtype=torch.float64)
Loss:
tensor(1.0794, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
55
Accuracy:
tensor(0.5875, dtype=torch.float64)
Loss:
tensor(1.0889, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
56
Accuracy:
tensor(0.5958, dtype=torch.float64)
Loss:
tensor(1.1275, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
57
Accuracy:
tensor(0.6500, dtype=torch.float64)
Loss:
tensor(1.0516, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
58
Accuracy:
tensor(0.5583, dtype=torch.float64)
Loss:
tensor(1.2064, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
59
Accuracy:
tensor(0.6458, dtype=torch.float64)
Loss:
tensor(1.0770, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
60
########################
Validation Accuracy:
tensor(0.2767, dtype=torch.float64)
########################
Accuracy:
tensor(0.6875, dtype=torch.float64)
Loss:
tensor(1.0061, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
61
Accuracy:
tensor(0.6333, dtype=torch.float64)
Loss:
tensor(1.0606, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
62
Accuracy:
tensor(0.6208, dtype=torch.float64)
Loss:
tensor(1.0521, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
63
Accuracy:
tensor(0.6375, dtype=torch.float64)
Loss:
tensor(1.1231, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
64
Accuracy:
tensor(0.6625, dtype=torch.float64)
Loss:
tensor(1.0410, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
65
Accuracy:
tensor(0.6375, dtype=torch.float64)
Loss:
tensor(1.0597, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
66
Accuracy:
tensor(0.6250, dtype=torch.float64)
Loss:
tensor(1.0466, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
67
Accuracy:
tensor(0.6375, dtype=torch.float64)
Loss:
tensor(1.1170, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
68
Accuracy:
tensor(0.6167, dtype=torch.float64)
Loss:
tensor(1.1727, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
69
Accuracy:
tensor(0.6333, dtype=torch.float64)
Loss:
tensor(1.1137, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
70
Accuracy:
tensor(0.6333, dtype=torch.float64)
Loss:
tensor(1.0831, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
71
Accuracy:
tensor(0.5917, dtype=torch.float64)
Loss:
tensor(1.2385, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
72
Accuracy:
tensor(0.5292, dtype=torch.float64)
Loss:
tensor(1.2381, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
73
Accuracy:
tensor(0.5958, dtype=torch.float64)
Loss:
tensor(1.1203, grad_fn=<NllLossBackward>)
Epoch:
7
Batch:
74
Accuracy:
tensor(0.7250, dtype=torch.float64)
Loss:
tensor(0.8651, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
0
########################
Validation Accuracy:
tensor(0.2833, dtype=torch.float64)
########################
Accuracy:
tensor(0.7083, dtype=torch.float64)
Loss:
tensor(0.9863, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
1
Accuracy:
tensor(0.7375, dtype=torch.float64)
Loss:
tensor(0.9065, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
2
Accuracy:
tensor(0.7583, dtype=torch.float64)
Loss:
tensor(0.9272, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
3
Accuracy:
tensor(0.7500, dtype=torch.float64)
Loss:
tensor(0.8740, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
4
Accuracy:
tensor(0.7583, dtype=torch.float64)
Loss:
tensor(0.7971, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
5
Accuracy:
tensor(0.7833, dtype=torch.float64)
Loss:
tensor(0.8168, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
6
Accuracy:
tensor(0.7167, dtype=torch.float64)
Loss:
tensor(0.9200, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
7
Accuracy:
tensor(0.7292, dtype=torch.float64)
Loss:
tensor(0.9071, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
8
Accuracy:
tensor(0.7292, dtype=torch.float64)
Loss:
tensor(0.8809, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
9
Accuracy:
tensor(0.7125, dtype=torch.float64)
Loss:
tensor(0.9405, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
10
Accuracy:
tensor(0.7458, dtype=torch.float64)
Loss:
tensor(0.8794, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
11
Accuracy:
tensor(0.7167, dtype=torch.float64)
Loss:
tensor(0.9178, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
12
Accuracy:
tensor(0.7417, dtype=torch.float64)
Loss:
tensor(0.8380, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
13
Accuracy:
tensor(0.7583, dtype=torch.float64)
Loss:
tensor(0.8772, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
14
Accuracy:
tensor(0.6958, dtype=torch.float64)
Loss:
tensor(0.9054, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
15
Accuracy:
tensor(0.7458, dtype=torch.float64)
Loss:
tensor(0.8899, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
16
Accuracy:
tensor(0.7292, dtype=torch.float64)
Loss:
tensor(0.9246, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
17
Accuracy:
tensor(0.7583, dtype=torch.float64)
Loss:
tensor(0.8660, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
18
Accuracy:
tensor(0.7500, dtype=torch.float64)
Loss:
tensor(0.8707, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
19
Accuracy:
tensor(0.6917, dtype=torch.float64)
Loss:
tensor(0.9322, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
20
########################
Validation Accuracy:
tensor(0.2933, dtype=torch.float64)
########################
Accuracy:
tensor(0.6833, dtype=torch.float64)
Loss:
tensor(0.9194, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
21
Accuracy:
tensor(0.7250, dtype=torch.float64)
Loss:
tensor(0.9372, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
22
Accuracy:
tensor(0.7375, dtype=torch.float64)
Loss:
tensor(0.8376, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
23
Accuracy:
tensor(0.7125, dtype=torch.float64)
Loss:
tensor(0.9060, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
24
Accuracy:
tensor(0.6875, dtype=torch.float64)
Loss:
tensor(0.9523, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
25
Accuracy:
tensor(0.7042, dtype=torch.float64)
Loss:
tensor(0.9498, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
26
Accuracy:
tensor(0.7750, dtype=torch.float64)
Loss:
tensor(0.8487, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
27
Accuracy:
tensor(0.7292, dtype=torch.float64)
Loss:
tensor(0.9466, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
28
Accuracy:
tensor(0.7000, dtype=torch.float64)
Loss:
tensor(0.9360, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
29
Accuracy:
tensor(0.7250, dtype=torch.float64)
Loss:
tensor(0.9399, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
30
Accuracy:
tensor(0.6917, dtype=torch.float64)
Loss:
tensor(0.9648, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
31
Accuracy:
tensor(0.6625, dtype=torch.float64)
Loss:
tensor(1.0007, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
32
Accuracy:
tensor(0.7125, dtype=torch.float64)
Loss:
tensor(0.9798, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
33
Accuracy:
tensor(0.7375, dtype=torch.float64)
Loss:
tensor(0.8812, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
34
Accuracy:
tensor(0.7458, dtype=torch.float64)
Loss:
tensor(0.8577, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
35
Accuracy:
tensor(0.6917, dtype=torch.float64)
Loss:
tensor(0.9625, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
36
Accuracy:
tensor(0.7375, dtype=torch.float64)
Loss:
tensor(0.8442, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
37
Accuracy:
tensor(0.7042, dtype=torch.float64)
Loss:
tensor(0.8787, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
38
Accuracy:
tensor(0.7042, dtype=torch.float64)
Loss:
tensor(0.9176, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
39
Accuracy:
tensor(0.6542, dtype=torch.float64)
Loss:
tensor(1.0118, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
40
########################
Validation Accuracy:
tensor(0.2933, dtype=torch.float64)
########################
Accuracy:
tensor(0.6958, dtype=torch.float64)
Loss:
tensor(0.9722, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
41
Accuracy:
tensor(0.7667, dtype=torch.float64)
Loss:
tensor(0.8578, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
42
Accuracy:
tensor(0.7167, dtype=torch.float64)
Loss:
tensor(0.8830, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
43
Accuracy:
tensor(0.7083, dtype=torch.float64)
Loss:
tensor(0.8874, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
44
Accuracy:
tensor(0.7667, dtype=torch.float64)
Loss:
tensor(0.8408, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
45
Accuracy:
tensor(0.7042, dtype=torch.float64)
Loss:
tensor(0.9206, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
46
Accuracy:
tensor(0.6750, dtype=torch.float64)
Loss:
tensor(0.9632, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
47
Accuracy:
tensor(0.6500, dtype=torch.float64)
Loss:
tensor(1.0432, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
48
Accuracy:
tensor(0.6958, dtype=torch.float64)
Loss:
tensor(0.9098, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
49
Accuracy:
tensor(0.6667, dtype=torch.float64)
Loss:
tensor(0.9605, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
50
Accuracy:
tensor(0.6792, dtype=torch.float64)
Loss:
tensor(0.9478, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
51
Accuracy:
tensor(0.7167, dtype=torch.float64)
Loss:
tensor(0.9246, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
52
Accuracy:
tensor(0.6625, dtype=torch.float64)
Loss:
tensor(1.0734, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
53
Accuracy:
tensor(0.7292, dtype=torch.float64)
Loss:
tensor(0.8854, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
54
Accuracy:
tensor(0.6958, dtype=torch.float64)
Loss:
tensor(0.9473, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
55
Accuracy:
tensor(0.6500, dtype=torch.float64)
Loss:
tensor(0.9772, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
56
Accuracy:
tensor(0.6667, dtype=torch.float64)
Loss:
tensor(0.9878, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
57
Accuracy:
tensor(0.6958, dtype=torch.float64)
Loss:
tensor(0.8800, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
58
Accuracy:
tensor(0.6917, dtype=torch.float64)
Loss:
tensor(0.9722, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
59
Accuracy:
tensor(0.6833, dtype=torch.float64)
Loss:
tensor(0.9226, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
60
########################
Validation Accuracy:
tensor(0.2750, dtype=torch.float64)
########################
Accuracy:
tensor(0.6833, dtype=torch.float64)
Loss:
tensor(0.9990, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
61
Accuracy:
tensor(0.7042, dtype=torch.float64)
Loss:
tensor(0.9056, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
62
Accuracy:
tensor(0.6458, dtype=torch.float64)
Loss:
tensor(0.9846, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
63
Accuracy:
tensor(0.6833, dtype=torch.float64)
Loss:
tensor(0.9244, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
64
Accuracy:
tensor(0.5083, dtype=torch.float64)
Loss:
tensor(1.2715, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
65
Accuracy:
tensor(0.6417, dtype=torch.float64)
Loss:
tensor(1.0521, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
66
Accuracy:
tensor(0.6667, dtype=torch.float64)
Loss:
tensor(1.0477, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
67
Accuracy:
tensor(0.6542, dtype=torch.float64)
Loss:
tensor(0.9937, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
68
Accuracy:
tensor(0.6917, dtype=torch.float64)
Loss:
tensor(0.8792, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
69
Accuracy:
tensor(0.5917, dtype=torch.float64)
Loss:
tensor(1.1871, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
70
Accuracy:
tensor(0.6375, dtype=torch.float64)
Loss:
tensor(1.0307, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
71
Accuracy:
tensor(0.6417, dtype=torch.float64)
Loss:
tensor(1.0674, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
72
Accuracy:
tensor(0.6708, dtype=torch.float64)
Loss:
tensor(0.9684, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
73
Accuracy:
tensor(0.6458, dtype=torch.float64)
Loss:
tensor(0.9858, grad_fn=<NllLossBackward>)
Epoch:
8
Batch:
74
Accuracy:
tensor(0.7708, dtype=torch.float64)
Loss:
tensor(0.7497, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
0
########################
Validation Accuracy:
tensor(0.2833, dtype=torch.float64)
########################
Accuracy:
tensor(0.7333, dtype=torch.float64)
Loss:
tensor(0.8393, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
1
Accuracy:
tensor(0.7458, dtype=torch.float64)
Loss:
tensor(0.7856, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
2
Accuracy:
tensor(0.7625, dtype=torch.float64)
Loss:
tensor(0.8321, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
3
Accuracy:
tensor(0.7417, dtype=torch.float64)
Loss:
tensor(0.7656, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
4
Accuracy:
tensor(0.7750, dtype=torch.float64)
Loss:
tensor(0.7718, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
5
Accuracy:
tensor(0.8083, dtype=torch.float64)
Loss:
tensor(0.7287, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
6
Accuracy:
tensor(0.7792, dtype=torch.float64)
Loss:
tensor(0.7587, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
7
Accuracy:
tensor(0.7708, dtype=torch.float64)
Loss:
tensor(0.7062, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
8
Accuracy:
tensor(0.7833, dtype=torch.float64)
Loss:
tensor(0.7148, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
9
Accuracy:
tensor(0.7917, dtype=torch.float64)
Loss:
tensor(0.7250, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
10
Accuracy:
tensor(0.7167, dtype=torch.float64)
Loss:
tensor(0.8866, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
11
Accuracy:
tensor(0.8125, dtype=torch.float64)
Loss:
tensor(0.7602, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
12
Accuracy:
tensor(0.7792, dtype=torch.float64)
Loss:
tensor(0.7622, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
13
Accuracy:
tensor(0.7958, dtype=torch.float64)
Loss:
tensor(0.7764, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
14
Accuracy:
tensor(0.7833, dtype=torch.float64)
Loss:
tensor(0.7180, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
15
Accuracy:
tensor(0.8167, dtype=torch.float64)
Loss:
tensor(0.7187, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
16
Accuracy:
tensor(0.7625, dtype=torch.float64)
Loss:
tensor(0.7926, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
17
Accuracy:
tensor(0.7625, dtype=torch.float64)
Loss:
tensor(0.7699, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
18
Accuracy:
tensor(0.7417, dtype=torch.float64)
Loss:
tensor(0.8250, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
19
Accuracy:
tensor(0.7375, dtype=torch.float64)
Loss:
tensor(0.7979, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
20
########################
Validation Accuracy:
tensor(0.2833, dtype=torch.float64)
########################
Accuracy:
tensor(0.8083, dtype=torch.float64)
Loss:
tensor(0.7145, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
21
Accuracy:
tensor(0.7583, dtype=torch.float64)
Loss:
tensor(0.7836, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
22
Accuracy:
tensor(0.7625, dtype=torch.float64)
Loss:
tensor(0.8098, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
23
Accuracy:
tensor(0.7458, dtype=torch.float64)
Loss:
tensor(0.7825, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
24
Accuracy:
tensor(0.7417, dtype=torch.float64)
Loss:
tensor(0.8310, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
25
Accuracy:
tensor(0.7208, dtype=torch.float64)
Loss:
tensor(0.7974, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
26
Accuracy:
tensor(0.7458, dtype=torch.float64)
Loss:
tensor(0.8019, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
27
Accuracy:
tensor(0.7708, dtype=torch.float64)
Loss:
tensor(0.8276, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
28
Accuracy:
tensor(0.7625, dtype=torch.float64)
Loss:
tensor(0.7835, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
29
Accuracy:
tensor(0.7833, dtype=torch.float64)
Loss:
tensor(0.7222, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
30
Accuracy:
tensor(0.7583, dtype=torch.float64)
Loss:
tensor(0.7634, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
31
Accuracy:
tensor(0.7625, dtype=torch.float64)
Loss:
tensor(0.8118, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
32
Accuracy:
tensor(0.7417, dtype=torch.float64)
Loss:
tensor(0.7456, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
33
Accuracy:
tensor(0.7625, dtype=torch.float64)
Loss:
tensor(0.7557, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
34
Accuracy:
tensor(0.7292, dtype=torch.float64)
Loss:
tensor(0.8136, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
35
Accuracy:
tensor(0.6792, dtype=torch.float64)
Loss:
tensor(0.8756, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
36
Accuracy:
tensor(0.7542, dtype=torch.float64)
Loss:
tensor(0.7755, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
37
Accuracy:
tensor(0.7333, dtype=torch.float64)
Loss:
tensor(0.8563, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
38
Accuracy:
tensor(0.7583, dtype=torch.float64)
Loss:
tensor(0.8047, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
39
Accuracy:
tensor(0.7625, dtype=torch.float64)
Loss:
tensor(0.8013, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
40
########################
Validation Accuracy:
tensor(0.2967, dtype=torch.float64)
########################
Accuracy:
tensor(0.7583, dtype=torch.float64)
Loss:
tensor(0.8388, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
41
Accuracy:
tensor(0.7750, dtype=torch.float64)
Loss:
tensor(0.7569, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
42
Accuracy:
tensor(0.7625, dtype=torch.float64)
Loss:
tensor(0.8087, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
43
Accuracy:
tensor(0.7375, dtype=torch.float64)
Loss:
tensor(0.8388, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
44
Accuracy:
tensor(0.7000, dtype=torch.float64)
Loss:
tensor(0.9158, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
45
Accuracy:
tensor(0.7708, dtype=torch.float64)
Loss:
tensor(0.7922, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
46
Accuracy:
tensor(0.7292, dtype=torch.float64)
Loss:
tensor(0.8386, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
47
Accuracy:
tensor(0.6833, dtype=torch.float64)
Loss:
tensor(0.9075, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
48
Accuracy:
tensor(0.7292, dtype=torch.float64)
Loss:
tensor(0.9386, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
49
Accuracy:
tensor(0.7125, dtype=torch.float64)
Loss:
tensor(0.8055, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
50
Accuracy:
tensor(0.7292, dtype=torch.float64)
Loss:
tensor(0.8742, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
51
Accuracy:
tensor(0.7125, dtype=torch.float64)
Loss:
tensor(0.8409, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
52
Accuracy:
tensor(0.7083, dtype=torch.float64)
Loss:
tensor(0.8427, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
53
Accuracy:
tensor(0.7292, dtype=torch.float64)
Loss:
tensor(0.8005, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
54
Accuracy:
tensor(0.6583, dtype=torch.float64)
Loss:
tensor(0.9712, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
55
Accuracy:
tensor(0.7375, dtype=torch.float64)
Loss:
tensor(0.8193, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
56
Accuracy:
tensor(0.7208, dtype=torch.float64)
Loss:
tensor(0.8314, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
57
Accuracy:
tensor(0.6958, dtype=torch.float64)
Loss:
tensor(0.8529, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
58
Accuracy:
tensor(0.7042, dtype=torch.float64)
Loss:
tensor(0.9152, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
59
Accuracy:
tensor(0.7125, dtype=torch.float64)
Loss:
tensor(0.8320, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
60
########################
Validation Accuracy:
tensor(0.2967, dtype=torch.float64)
########################
Accuracy:
tensor(0.7417, dtype=torch.float64)
Loss:
tensor(0.8937, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
61
Accuracy:
tensor(0.7292, dtype=torch.float64)
Loss:
tensor(0.8648, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
62
Accuracy:
tensor(0.7375, dtype=torch.float64)
Loss:
tensor(0.8261, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
63
Accuracy:
tensor(0.7042, dtype=torch.float64)
Loss:
tensor(0.8709, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
64
Accuracy:
tensor(0.7125, dtype=torch.float64)
Loss:
tensor(0.9179, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
65
Accuracy:
tensor(0.7292, dtype=torch.float64)
Loss:
tensor(0.8446, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
66
Accuracy:
tensor(0.7375, dtype=torch.float64)
Loss:
tensor(0.7535, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
67
Accuracy:
tensor(0.6542, dtype=torch.float64)
Loss:
tensor(0.9717, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
68
Accuracy:
tensor(0.6958, dtype=torch.float64)
Loss:
tensor(0.9120, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
69
Accuracy:
tensor(0.7875, dtype=torch.float64)
Loss:
tensor(0.7689, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
70
Accuracy:
tensor(0.7292, dtype=torch.float64)
Loss:
tensor(0.8737, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
71
Accuracy:
tensor(0.7292, dtype=torch.float64)
Loss:
tensor(0.8299, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
72
Accuracy:
tensor(0.6917, dtype=torch.float64)
Loss:
tensor(0.9267, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
73
Accuracy:
tensor(0.7083, dtype=torch.float64)
Loss:
tensor(0.8579, grad_fn=<NllLossBackward>)
Epoch:
9
Batch:
74
Accuracy:
tensor(0.8583, dtype=torch.float64)
Loss:
tensor(0.5667, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
0
########################
Validation Accuracy:
tensor(0.3017, dtype=torch.float64)
########################
Accuracy:
tensor(0.8375, dtype=torch.float64)
Loss:
tensor(0.6219, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
1
Accuracy:
tensor(0.8167, dtype=torch.float64)
Loss:
tensor(0.7109, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
2
Accuracy:
tensor(0.7958, dtype=torch.float64)
Loss:
tensor(0.7255, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
3
Accuracy:
tensor(0.7917, dtype=torch.float64)
Loss:
tensor(0.6890, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
4
Accuracy:
tensor(0.8417, dtype=torch.float64)
Loss:
tensor(0.6268, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
5
Accuracy:
tensor(0.7833, dtype=torch.float64)
Loss:
tensor(0.6667, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
6
Accuracy:
tensor(0.8375, dtype=torch.float64)
Loss:
tensor(0.6115, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
7
Accuracy:
tensor(0.8083, dtype=torch.float64)
Loss:
tensor(0.6925, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
8
Accuracy:
tensor(0.8208, dtype=torch.float64)
Loss:
tensor(0.6730, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
9
Accuracy:
tensor(0.7625, dtype=torch.float64)
Loss:
tensor(0.7254, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
10
Accuracy:
tensor(0.8208, dtype=torch.float64)
Loss:
tensor(0.6856, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
11
Accuracy:
tensor(0.8417, dtype=torch.float64)
Loss:
tensor(0.6263, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
12
Accuracy:
tensor(0.7667, dtype=torch.float64)
Loss:
tensor(0.7114, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
13
Accuracy:
tensor(0.8500, dtype=torch.float64)
Loss:
tensor(0.6869, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
14
Accuracy:
tensor(0.7375, dtype=torch.float64)
Loss:
tensor(0.8089, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
15
Accuracy:
tensor(0.8042, dtype=torch.float64)
Loss:
tensor(0.6632, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
16
Accuracy:
tensor(0.8500, dtype=torch.float64)
Loss:
tensor(0.6348, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
17
Accuracy:
tensor(0.7833, dtype=torch.float64)
Loss:
tensor(0.7166, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
18
Accuracy:
tensor(0.8083, dtype=torch.float64)
Loss:
tensor(0.6053, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
19
Accuracy:
tensor(0.8583, dtype=torch.float64)
Loss:
tensor(0.6003, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
20
########################
Validation Accuracy:
tensor(0.2900, dtype=torch.float64)
########################
Accuracy:
tensor(0.7500, dtype=torch.float64)
Loss:
tensor(0.7407, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
21
Accuracy:
tensor(0.8583, dtype=torch.float64)
Loss:
tensor(0.5887, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
22
Accuracy:
tensor(0.7833, dtype=torch.float64)
Loss:
tensor(0.7189, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
23
Accuracy:
tensor(0.8417, dtype=torch.float64)
Loss:
tensor(0.6311, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
24
Accuracy:
tensor(0.7875, dtype=torch.float64)
Loss:
tensor(0.7095, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
25
Accuracy:
tensor(0.7458, dtype=torch.float64)
Loss:
tensor(0.8170, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
26
Accuracy:
tensor(0.8417, dtype=torch.float64)
Loss:
tensor(0.5944, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
27
Accuracy:
tensor(0.7958, dtype=torch.float64)
Loss:
tensor(0.7439, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
28
Accuracy:
tensor(0.7375, dtype=torch.float64)
Loss:
tensor(0.7744, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
29
Accuracy:
tensor(0.8000, dtype=torch.float64)
Loss:
tensor(0.6935, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
30
Accuracy:
tensor(0.8042, dtype=torch.float64)
Loss:
tensor(0.6395, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
31
Accuracy:
tensor(0.8042, dtype=torch.float64)
Loss:
tensor(0.6944, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
32
Accuracy:
tensor(0.7333, dtype=torch.float64)
Loss:
tensor(0.8016, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
33
Accuracy:
tensor(0.8375, dtype=torch.float64)
Loss:
tensor(0.6141, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
34
Accuracy:
tensor(0.8000, dtype=torch.float64)
Loss:
tensor(0.6518, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
35
Accuracy:
tensor(0.7958, dtype=torch.float64)
Loss:
tensor(0.6918, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
36
Accuracy:
tensor(0.7542, dtype=torch.float64)
Loss:
tensor(0.7319, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
37
Accuracy:
tensor(0.7375, dtype=torch.float64)
Loss:
tensor(0.7706, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
38
Accuracy:
tensor(0.7667, dtype=torch.float64)
Loss:
tensor(0.7247, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
39
Accuracy:
tensor(0.8208, dtype=torch.float64)
Loss:
tensor(0.6556, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
40
########################
Validation Accuracy:
tensor(0.2900, dtype=torch.float64)
########################
Accuracy:
tensor(0.8167, dtype=torch.float64)
Loss:
tensor(0.6935, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
41
Accuracy:
tensor(0.8000, dtype=torch.float64)
Loss:
tensor(0.6795, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
42
Accuracy:
tensor(0.8083, dtype=torch.float64)
Loss:
tensor(0.6782, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
43
Accuracy:
tensor(0.7542, dtype=torch.float64)
Loss:
tensor(0.7101, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
44
Accuracy:
tensor(0.7292, dtype=torch.float64)
Loss:
tensor(0.7685, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
45
Accuracy:
tensor(0.7792, dtype=torch.float64)
Loss:
tensor(0.7085, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
46
Accuracy:
tensor(0.7458, dtype=torch.float64)
Loss:
tensor(0.7697, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
47
Accuracy:
tensor(0.6917, dtype=torch.float64)
Loss:
tensor(0.8240, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
48
Accuracy:
tensor(0.7917, dtype=torch.float64)
Loss:
tensor(0.7247, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
49
Accuracy:
tensor(0.7458, dtype=torch.float64)
Loss:
tensor(0.8178, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
50
Accuracy:
tensor(0.6792, dtype=torch.float64)
Loss:
tensor(0.8855, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
51
Accuracy:
tensor(0.7542, dtype=torch.float64)
Loss:
tensor(0.7397, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
52
Accuracy:
tensor(0.7833, dtype=torch.float64)
Loss:
tensor(0.7549, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
53
Accuracy:
tensor(0.7542, dtype=torch.float64)
Loss:
tensor(0.7769, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
54
Accuracy:
tensor(0.7125, dtype=torch.float64)
Loss:
tensor(0.8155, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
55
Accuracy:
tensor(0.7667, dtype=torch.float64)
Loss:
tensor(0.7698, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
56
Accuracy:
tensor(0.7583, dtype=torch.float64)
Loss:
tensor(0.7850, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
57
Accuracy:
tensor(0.7542, dtype=torch.float64)
Loss:
tensor(0.7535, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
58
Accuracy:
tensor(0.7583, dtype=torch.float64)
Loss:
tensor(0.7516, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
59
Accuracy:
tensor(0.7583, dtype=torch.float64)
Loss:
tensor(0.6907, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
60
########################
Validation Accuracy:
tensor(0.2817, dtype=torch.float64)
########################
Accuracy:
tensor(0.7875, dtype=torch.float64)
Loss:
tensor(0.7366, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
61
Accuracy:
tensor(0.7250, dtype=torch.float64)
Loss:
tensor(0.7900, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
62
Accuracy:
tensor(0.7125, dtype=torch.float64)
Loss:
tensor(0.8154, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
63
Accuracy:
tensor(0.7333, dtype=torch.float64)
Loss:
tensor(0.8087, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
64
Accuracy:
tensor(0.7417, dtype=torch.float64)
Loss:
tensor(0.8614, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
65
Accuracy:
tensor(0.7917, dtype=torch.float64)
Loss:
tensor(0.7001, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
66
Accuracy:
tensor(0.7792, dtype=torch.float64)
Loss:
tensor(0.7207, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
67
Accuracy:
tensor(0.7458, dtype=torch.float64)
Loss:
tensor(0.7584, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
68
Accuracy:
tensor(0.7750, dtype=torch.float64)
Loss:
tensor(0.7485, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
69
Accuracy:
tensor(0.7125, dtype=torch.float64)
Loss:
tensor(0.8391, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
70
Accuracy:
tensor(0.7458, dtype=torch.float64)
Loss:
tensor(0.7778, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
71
Accuracy:
tensor(0.7125, dtype=torch.float64)
Loss:
tensor(0.7948, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
72
Accuracy:
tensor(0.7292, dtype=torch.float64)
Loss:
tensor(0.7625, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
73
Accuracy:
tensor(0.7458, dtype=torch.float64)
Loss:
tensor(0.7231, grad_fn=<NllLossBackward>)
Epoch:
10
Batch:
74
Accuracy:
tensor(0.8417, dtype=torch.float64)
Loss:
tensor(0.6158, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
0
########################
Validation Accuracy:
tensor(0.2883, dtype=torch.float64)
########################
Accuracy:
tensor(0.8583, dtype=torch.float64)
Loss:
tensor(0.5442, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
1
Accuracy:
tensor(0.8167, dtype=torch.float64)
Loss:
tensor(0.5869, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
2
Accuracy:
tensor(0.8375, dtype=torch.float64)
Loss:
tensor(0.5720, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
3
Accuracy:
tensor(0.8375, dtype=torch.float64)
Loss:
tensor(0.5798, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
4
Accuracy:
tensor(0.8708, dtype=torch.float64)
Loss:
tensor(0.5281, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
5
Accuracy:
tensor(0.8083, dtype=torch.float64)
Loss:
tensor(0.6512, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
6
Accuracy:
tensor(0.8375, dtype=torch.float64)
Loss:
tensor(0.5819, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
7
Accuracy:
tensor(0.8500, dtype=torch.float64)
Loss:
tensor(0.5857, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
8
Accuracy:
tensor(0.8458, dtype=torch.float64)
Loss:
tensor(0.5547, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
9
Accuracy:
tensor(0.9042, dtype=torch.float64)
Loss:
tensor(0.4606, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
10
Accuracy:
tensor(0.8417, dtype=torch.float64)
Loss:
tensor(0.6133, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
11
Accuracy:
tensor(0.8417, dtype=torch.float64)
Loss:
tensor(0.5807, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
12
Accuracy:
tensor(0.8375, dtype=torch.float64)
Loss:
tensor(0.5209, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
13
Accuracy:
tensor(0.8625, dtype=torch.float64)
Loss:
tensor(0.5446, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
14
Accuracy:
tensor(0.8167, dtype=torch.float64)
Loss:
tensor(0.5926, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
15
Accuracy:
tensor(0.8000, dtype=torch.float64)
Loss:
tensor(0.5938, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
16
Accuracy:
tensor(0.8000, dtype=torch.float64)
Loss:
tensor(0.6613, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
17
Accuracy:
tensor(0.8500, dtype=torch.float64)
Loss:
tensor(0.5231, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
18
Accuracy:
tensor(0.8500, dtype=torch.float64)
Loss:
tensor(0.5737, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
19
Accuracy:
tensor(0.7667, dtype=torch.float64)
Loss:
tensor(0.6575, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
20
########################
Validation Accuracy:
tensor(0.2767, dtype=torch.float64)
########################
Accuracy:
tensor(0.8125, dtype=torch.float64)
Loss:
tensor(0.5836, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
21
Accuracy:
tensor(0.8375, dtype=torch.float64)
Loss:
tensor(0.5861, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
22
Accuracy:
tensor(0.8208, dtype=torch.float64)
Loss:
tensor(0.6106, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
23
Accuracy:
tensor(0.8625, dtype=torch.float64)
Loss:
tensor(0.5515, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
24
Accuracy:
tensor(0.8333, dtype=torch.float64)
Loss:
tensor(0.5980, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
25
Accuracy:
tensor(0.8083, dtype=torch.float64)
Loss:
tensor(0.6210, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
26
Accuracy:
tensor(0.8042, dtype=torch.float64)
Loss:
tensor(0.6877, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
27
Accuracy:
tensor(0.8167, dtype=torch.float64)
Loss:
tensor(0.5801, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
28
Accuracy:
tensor(0.7792, dtype=torch.float64)
Loss:
tensor(0.6190, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
29
Accuracy:
tensor(0.8458, dtype=torch.float64)
Loss:
tensor(0.5189, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
30
Accuracy:
tensor(0.8125, dtype=torch.float64)
Loss:
tensor(0.6627, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
31
Accuracy:
tensor(0.7833, dtype=torch.float64)
Loss:
tensor(0.6958, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
32
Accuracy:
tensor(0.7667, dtype=torch.float64)
Loss:
tensor(0.6856, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
33
Accuracy:
tensor(0.8208, dtype=torch.float64)
Loss:
tensor(0.6191, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
34
Accuracy:
tensor(0.8250, dtype=torch.float64)
Loss:
tensor(0.5984, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
35
Accuracy:
tensor(0.8250, dtype=torch.float64)
Loss:
tensor(0.5731, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
36
Accuracy:
tensor(0.8208, dtype=torch.float64)
Loss:
tensor(0.5946, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
37
Accuracy:
tensor(0.7958, dtype=torch.float64)
Loss:
tensor(0.6205, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
38
Accuracy:
tensor(0.8292, dtype=torch.float64)
Loss:
tensor(0.6089, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
39
Accuracy:
tensor(0.8000, dtype=torch.float64)
Loss:
tensor(0.6461, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
40
########################
Validation Accuracy:
tensor(0.2983, dtype=torch.float64)
########################
Accuracy:
tensor(0.7667, dtype=torch.float64)
Loss:
tensor(0.7045, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
41
Accuracy:
tensor(0.8208, dtype=torch.float64)
Loss:
tensor(0.6249, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
42
Accuracy:
tensor(0.7875, dtype=torch.float64)
Loss:
tensor(0.6331, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
43
Accuracy:
tensor(0.7875, dtype=torch.float64)
Loss:
tensor(0.6130, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
44
Accuracy:
tensor(0.7500, dtype=torch.float64)
Loss:
tensor(0.6596, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
45
Accuracy:
tensor(0.7750, dtype=torch.float64)
Loss:
tensor(0.7195, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
46
Accuracy:
tensor(0.7833, dtype=torch.float64)
Loss:
tensor(0.6540, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
47
Accuracy:
tensor(0.8083, dtype=torch.float64)
Loss:
tensor(0.6491, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
48
Accuracy:
tensor(0.7833, dtype=torch.float64)
Loss:
tensor(0.6774, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
49
Accuracy:
tensor(0.7625, dtype=torch.float64)
Loss:
tensor(0.6773, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
50
Accuracy:
tensor(0.7667, dtype=torch.float64)
Loss:
tensor(0.7195, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
51
Accuracy:
tensor(0.7667, dtype=torch.float64)
Loss:
tensor(0.6837, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
52
Accuracy:
tensor(0.8000, dtype=torch.float64)
Loss:
tensor(0.6772, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
53
Accuracy:
tensor(0.7750, dtype=torch.float64)
Loss:
tensor(0.6869, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
54
Accuracy:
tensor(0.7833, dtype=torch.float64)
Loss:
tensor(0.6340, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
55
Accuracy:
tensor(0.7625, dtype=torch.float64)
Loss:
tensor(0.7194, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
56
Accuracy:
tensor(0.7833, dtype=torch.float64)
Loss:
tensor(0.6328, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
57
Accuracy:
tensor(0.7708, dtype=torch.float64)
Loss:
tensor(0.7050, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
58
Accuracy:
tensor(0.7792, dtype=torch.float64)
Loss:
tensor(0.6683, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
59
Accuracy:
tensor(0.8083, dtype=torch.float64)
Loss:
tensor(0.6192, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
60
########################
Validation Accuracy:
tensor(0.2850, dtype=torch.float64)
########################
Accuracy:
tensor(0.7292, dtype=torch.float64)
Loss:
tensor(0.7279, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
61
Accuracy:
tensor(0.7708, dtype=torch.float64)
Loss:
tensor(0.7313, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
62
Accuracy:
tensor(0.8000, dtype=torch.float64)
Loss:
tensor(0.6311, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
63
Accuracy:
tensor(0.7542, dtype=torch.float64)
Loss:
tensor(0.7051, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
64
Accuracy:
tensor(0.8292, dtype=torch.float64)
Loss:
tensor(0.5851, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
65
Accuracy:
tensor(0.8042, dtype=torch.float64)
Loss:
tensor(0.6626, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
66
Accuracy:
tensor(0.8042, dtype=torch.float64)
Loss:
tensor(0.5976, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
67
Accuracy:
tensor(0.8042, dtype=torch.float64)
Loss:
tensor(0.6155, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
68
Accuracy:
tensor(0.7958, dtype=torch.float64)
Loss:
tensor(0.6831, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
69
Accuracy:
tensor(0.7750, dtype=torch.float64)
Loss:
tensor(0.7433, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
70
Accuracy:
tensor(0.7750, dtype=torch.float64)
Loss:
tensor(0.6526, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
71
Accuracy:
tensor(0.7000, dtype=torch.float64)
Loss:
tensor(0.8055, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
72
Accuracy:
tensor(0.7667, dtype=torch.float64)
Loss:
tensor(0.7117, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
73
Accuracy:
tensor(0.7500, dtype=torch.float64)
Loss:
tensor(0.7172, grad_fn=<NllLossBackward>)
Epoch:
11
Batch:
74
Accuracy:
tensor(0.8167, dtype=torch.float64)
Loss:
tensor(0.5576, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
0
########################
Validation Accuracy:
tensor(0.3000, dtype=torch.float64)
########################
Accuracy:
tensor(0.8500, dtype=torch.float64)
Loss:
tensor(0.5207, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
1
Accuracy:
tensor(0.8792, dtype=torch.float64)
Loss:
tensor(0.4537, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
2
Accuracy:
tensor(0.8708, dtype=torch.float64)
Loss:
tensor(0.4711, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
3
Accuracy:
tensor(0.8917, dtype=torch.float64)
Loss:
tensor(0.4690, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
4
Accuracy:
tensor(0.8958, dtype=torch.float64)
Loss:
tensor(0.4522, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
5
Accuracy:
tensor(0.8833, dtype=torch.float64)
Loss:
tensor(0.4423, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
6
Accuracy:
tensor(0.8542, dtype=torch.float64)
Loss:
tensor(0.5192, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
7
Accuracy:
tensor(0.8417, dtype=torch.float64)
Loss:
tensor(0.5343, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
8
Accuracy:
tensor(0.8625, dtype=torch.float64)
Loss:
tensor(0.5059, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
9
Accuracy:
tensor(0.8375, dtype=torch.float64)
Loss:
tensor(0.5385, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
10
Accuracy:
tensor(0.8917, dtype=torch.float64)
Loss:
tensor(0.4492, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
11
Accuracy:
tensor(0.8792, dtype=torch.float64)
Loss:
tensor(0.4969, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
12
Accuracy:
tensor(0.8458, dtype=torch.float64)
Loss:
tensor(0.5505, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
13
Accuracy:
tensor(0.8833, dtype=torch.float64)
Loss:
tensor(0.4826, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
14
Accuracy:
tensor(0.8542, dtype=torch.float64)
Loss:
tensor(0.5013, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
15
Accuracy:
tensor(0.8667, dtype=torch.float64)
Loss:
tensor(0.4803, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
16
Accuracy:
tensor(0.8667, dtype=torch.float64)
Loss:
tensor(0.4835, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
17
Accuracy:
tensor(0.8333, dtype=torch.float64)
Loss:
tensor(0.5652, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
18
Accuracy:
tensor(0.8667, dtype=torch.float64)
Loss:
tensor(0.5124, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
19
Accuracy:
tensor(0.8792, dtype=torch.float64)
Loss:
tensor(0.4774, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
20
########################
Validation Accuracy:
tensor(0.2933, dtype=torch.float64)
########################
Accuracy:
tensor(0.8542, dtype=torch.float64)
Loss:
tensor(0.5061, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
21
Accuracy:
tensor(0.8458, dtype=torch.float64)
Loss:
tensor(0.5291, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
22
Accuracy:
tensor(0.8833, dtype=torch.float64)
Loss:
tensor(0.4723, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
23
Accuracy:
tensor(0.8625, dtype=torch.float64)
Loss:
tensor(0.5315, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
24
Accuracy:
tensor(0.8458, dtype=torch.float64)
Loss:
tensor(0.5496, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
25
Accuracy:
tensor(0.8583, dtype=torch.float64)
Loss:
tensor(0.5823, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
26
Accuracy:
tensor(0.8833, dtype=torch.float64)
Loss:
tensor(0.4391, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
27
Accuracy:
tensor(0.8750, dtype=torch.float64)
Loss:
tensor(0.4599, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
28
Accuracy:
tensor(0.8375, dtype=torch.float64)
Loss:
tensor(0.5500, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
29
Accuracy:
tensor(0.8000, dtype=torch.float64)
Loss:
tensor(0.5869, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
30
Accuracy:
tensor(0.8042, dtype=torch.float64)
Loss:
tensor(0.5654, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
31
Accuracy:
tensor(0.8417, dtype=torch.float64)
Loss:
tensor(0.5324, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
32
Accuracy:
tensor(0.8250, dtype=torch.float64)
Loss:
tensor(0.5339, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
33
Accuracy:
tensor(0.8375, dtype=torch.float64)
Loss:
tensor(0.5621, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
34
Accuracy:
tensor(0.8583, dtype=torch.float64)
Loss:
tensor(0.5290, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
35
Accuracy:
tensor(0.8250, dtype=torch.float64)
Loss:
tensor(0.6015, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
36
Accuracy:
tensor(0.8083, dtype=torch.float64)
Loss:
tensor(0.6425, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
37
Accuracy:
tensor(0.8167, dtype=torch.float64)
Loss:
tensor(0.5670, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
38
Accuracy:
tensor(0.8167, dtype=torch.float64)
Loss:
tensor(0.6071, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
39
Accuracy:
tensor(0.8000, dtype=torch.float64)
Loss:
tensor(0.6254, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
40
########################
Validation Accuracy:
tensor(0.2883, dtype=torch.float64)
########################
Accuracy:
tensor(0.8333, dtype=torch.float64)
Loss:
tensor(0.5623, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
41
Accuracy:
tensor(0.8167, dtype=torch.float64)
Loss:
tensor(0.6487, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
42
Accuracy:
tensor(0.8667, dtype=torch.float64)
Loss:
tensor(0.5002, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
43
Accuracy:
tensor(0.8042, dtype=torch.float64)
Loss:
tensor(0.6061, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
44
Accuracy:
tensor(0.7625, dtype=torch.float64)
Loss:
tensor(0.6760, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
45
Accuracy:
tensor(0.8083, dtype=torch.float64)
Loss:
tensor(0.6016, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
46
Accuracy:
tensor(0.8083, dtype=torch.float64)
Loss:
tensor(0.6116, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
47
Accuracy:
tensor(0.7875, dtype=torch.float64)
Loss:
tensor(0.6894, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
48
Accuracy:
tensor(0.8583, dtype=torch.float64)
Loss:
tensor(0.4786, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
49
Accuracy:
tensor(0.7875, dtype=torch.float64)
Loss:
tensor(0.6278, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
50
Accuracy:
tensor(0.7958, dtype=torch.float64)
Loss:
tensor(0.5973, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
51
Accuracy:
tensor(0.8375, dtype=torch.float64)
Loss:
tensor(0.5465, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
52
Accuracy:
tensor(0.8625, dtype=torch.float64)
Loss:
tensor(0.5028, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
53
Accuracy:
tensor(0.8333, dtype=torch.float64)
Loss:
tensor(0.5437, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
54
Accuracy:
tensor(0.7667, dtype=torch.float64)
Loss:
tensor(0.6051, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
55
Accuracy:
tensor(0.8250, dtype=torch.float64)
Loss:
tensor(0.5790, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
56
Accuracy:
tensor(0.8292, dtype=torch.float64)
Loss:
tensor(0.5703, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
57
Accuracy:
tensor(0.8208, dtype=torch.float64)
Loss:
tensor(0.5682, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
58
Accuracy:
tensor(0.8167, dtype=torch.float64)
Loss:
tensor(0.5930, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
59
Accuracy:
tensor(0.7958, dtype=torch.float64)
Loss:
tensor(0.6214, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
60
########################
Validation Accuracy:
tensor(0.2867, dtype=torch.float64)
########################
Accuracy:
tensor(0.7792, dtype=torch.float64)
Loss:
tensor(0.6508, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
61
Accuracy:
tensor(0.7667, dtype=torch.float64)
Loss:
tensor(0.6393, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
62
Accuracy:
tensor(0.8167, dtype=torch.float64)
Loss:
tensor(0.5645, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
63
Accuracy:
tensor(0.7833, dtype=torch.float64)
Loss:
tensor(0.6234, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
64
Accuracy:
tensor(0.7875, dtype=torch.float64)
Loss:
tensor(0.6466, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
65
Accuracy:
tensor(0.8042, dtype=torch.float64)
Loss:
tensor(0.5446, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
66
Accuracy:
tensor(0.7833, dtype=torch.float64)
Loss:
tensor(0.6505, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
67
Accuracy:
tensor(0.7875, dtype=torch.float64)
Loss:
tensor(0.6417, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
68
Accuracy:
tensor(0.7500, dtype=torch.float64)
Loss:
tensor(0.6577, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
69
Accuracy:
tensor(0.8167, dtype=torch.float64)
Loss:
tensor(0.6146, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
70
Accuracy:
tensor(0.8083, dtype=torch.float64)
Loss:
tensor(0.6393, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
71
Accuracy:
tensor(0.7625, dtype=torch.float64)
Loss:
tensor(0.7045, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
72
Accuracy:
tensor(0.7875, dtype=torch.float64)
Loss:
tensor(0.6561, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
73
Accuracy:
tensor(0.7417, dtype=torch.float64)
Loss:
tensor(0.7415, grad_fn=<NllLossBackward>)
Epoch:
12
Batch:
74
Accuracy:
tensor(0.8833, dtype=torch.float64)
Loss:
tensor(0.4544, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
0
########################
Validation Accuracy:
tensor(0.2833, dtype=torch.float64)
########################
Accuracy:
tensor(0.9083, dtype=torch.float64)
Loss:
tensor(0.3766, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
1
Accuracy:
tensor(0.8750, dtype=torch.float64)
Loss:
tensor(0.4905, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
2
Accuracy:
tensor(0.9208, dtype=torch.float64)
Loss:
tensor(0.4200, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
3
Accuracy:
tensor(0.8500, dtype=torch.float64)
Loss:
tensor(0.4828, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
4
Accuracy:
tensor(0.8542, dtype=torch.float64)
Loss:
tensor(0.4800, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
5
Accuracy:
tensor(0.9042, dtype=torch.float64)
Loss:
tensor(0.4186, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
6
Accuracy:
tensor(0.8958, dtype=torch.float64)
Loss:
tensor(0.4387, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
7
Accuracy:
tensor(0.8833, dtype=torch.float64)
Loss:
tensor(0.4108, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
8
Accuracy:
tensor(0.8542, dtype=torch.float64)
Loss:
tensor(0.4817, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
9
Accuracy:
tensor(0.8667, dtype=torch.float64)
Loss:
tensor(0.4944, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
10
Accuracy:
tensor(0.8583, dtype=torch.float64)
Loss:
tensor(0.4190, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
11
Accuracy:
tensor(0.9375, dtype=torch.float64)
Loss:
tensor(0.3199, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
12
Accuracy:
tensor(0.9000, dtype=torch.float64)
Loss:
tensor(0.4007, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
13
Accuracy:
tensor(0.8917, dtype=torch.float64)
Loss:
tensor(0.4345, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
14
Accuracy:
tensor(0.8625, dtype=torch.float64)
Loss:
tensor(0.4414, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
15
Accuracy:
tensor(0.8417, dtype=torch.float64)
Loss:
tensor(0.4860, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
16
Accuracy:
tensor(0.8917, dtype=torch.float64)
Loss:
tensor(0.4058, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
17
Accuracy:
tensor(0.8792, dtype=torch.float64)
Loss:
tensor(0.4101, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
18
Accuracy:
tensor(0.8625, dtype=torch.float64)
Loss:
tensor(0.4676, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
19
Accuracy:
tensor(0.8917, dtype=torch.float64)
Loss:
tensor(0.4460, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
20
########################
Validation Accuracy:
tensor(0.2850, dtype=torch.float64)
########################
Accuracy:
tensor(0.8792, dtype=torch.float64)
Loss:
tensor(0.4662, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
21
Accuracy:
tensor(0.8750, dtype=torch.float64)
Loss:
tensor(0.4450, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
22
Accuracy:
tensor(0.8667, dtype=torch.float64)
Loss:
tensor(0.4631, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
23
Accuracy:
tensor(0.8958, dtype=torch.float64)
Loss:
tensor(0.4004, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
24
Accuracy:
tensor(0.8875, dtype=torch.float64)
Loss:
tensor(0.4376, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
25
Accuracy:
tensor(0.8750, dtype=torch.float64)
Loss:
tensor(0.4975, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
26
Accuracy:
tensor(0.8750, dtype=torch.float64)
Loss:
tensor(0.4834, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
27
Accuracy:
tensor(0.8583, dtype=torch.float64)
Loss:
tensor(0.5226, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
28
Accuracy:
tensor(0.8750, dtype=torch.float64)
Loss:
tensor(0.4118, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
29
Accuracy:
tensor(0.8375, dtype=torch.float64)
Loss:
tensor(0.5110, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
30
Accuracy:
tensor(0.9000, dtype=torch.float64)
Loss:
tensor(0.4354, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
31
Accuracy:
tensor(0.8500, dtype=torch.float64)
Loss:
tensor(0.4873, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
32
Accuracy:
tensor(0.8917, dtype=torch.float64)
Loss:
tensor(0.4323, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
33
Accuracy:
tensor(0.8750, dtype=torch.float64)
Loss:
tensor(0.4351, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
34
Accuracy:
tensor(0.8375, dtype=torch.float64)
Loss:
tensor(0.5119, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
35
Accuracy:
tensor(0.8250, dtype=torch.float64)
Loss:
tensor(0.5387, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
36
Accuracy:
tensor(0.8708, dtype=torch.float64)
Loss:
tensor(0.4304, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
37
Accuracy:
tensor(0.8250, dtype=torch.float64)
Loss:
tensor(0.5342, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
38
Accuracy:
tensor(0.8667, dtype=torch.float64)
Loss:
tensor(0.5070, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
39
Accuracy:
tensor(0.8750, dtype=torch.float64)
Loss:
tensor(0.4637, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
40
########################
Validation Accuracy:
tensor(0.2733, dtype=torch.float64)
########################
Accuracy:
tensor(0.8500, dtype=torch.float64)
Loss:
tensor(0.5334, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
41
Accuracy:
tensor(0.8458, dtype=torch.float64)
Loss:
tensor(0.4957, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
42
Accuracy:
tensor(0.8458, dtype=torch.float64)
Loss:
tensor(0.4687, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
43
Accuracy:
tensor(0.8167, dtype=torch.float64)
Loss:
tensor(0.5770, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
44
Accuracy:
tensor(0.8542, dtype=torch.float64)
Loss:
tensor(0.5156, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
45
Accuracy:
tensor(0.8583, dtype=torch.float64)
Loss:
tensor(0.4818, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
46
Accuracy:
tensor(0.8500, dtype=torch.float64)
Loss:
tensor(0.4609, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
47
Accuracy:
tensor(0.7708, dtype=torch.float64)
Loss:
tensor(0.6995, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
48
Accuracy:
tensor(0.8042, dtype=torch.float64)
Loss:
tensor(0.5568, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
49
Accuracy:
tensor(0.8875, dtype=torch.float64)
Loss:
tensor(0.4167, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
50
Accuracy:
tensor(0.8833, dtype=torch.float64)
Loss:
tensor(0.4459, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
51
Accuracy:
tensor(0.8708, dtype=torch.float64)
Loss:
tensor(0.4670, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
52
Accuracy:
tensor(0.8667, dtype=torch.float64)
Loss:
tensor(0.4707, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
53
Accuracy:
tensor(0.8250, dtype=torch.float64)
Loss:
tensor(0.5515, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
54
Accuracy:
tensor(0.8792, dtype=torch.float64)
Loss:
tensor(0.4554, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
55
Accuracy:
tensor(0.7917, dtype=torch.float64)
Loss:
tensor(0.6020, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
56
Accuracy:
tensor(0.8208, dtype=torch.float64)
Loss:
tensor(0.5821, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
57
Accuracy:
tensor(0.8542, dtype=torch.float64)
Loss:
tensor(0.4779, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
58
Accuracy:
tensor(0.8417, dtype=torch.float64)
Loss:
tensor(0.4696, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
59
Accuracy:
tensor(0.8917, dtype=torch.float64)
Loss:
tensor(0.4354, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
60
########################
Validation Accuracy:
tensor(0.2933, dtype=torch.float64)
########################
Accuracy:
tensor(0.8583, dtype=torch.float64)
Loss:
tensor(0.4746, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
61
Accuracy:
tensor(0.8208, dtype=torch.float64)
Loss:
tensor(0.5932, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
62
Accuracy:
tensor(0.8458, dtype=torch.float64)
Loss:
tensor(0.4601, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
63
Accuracy:
tensor(0.8292, dtype=torch.float64)
Loss:
tensor(0.4987, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
64
Accuracy:
tensor(0.8167, dtype=torch.float64)
Loss:
tensor(0.5910, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
65
Accuracy:
tensor(0.8083, dtype=torch.float64)
Loss:
tensor(0.5533, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
66
Accuracy:
tensor(0.8167, dtype=torch.float64)
Loss:
tensor(0.5265, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
67
Accuracy:
tensor(0.8333, dtype=torch.float64)
Loss:
tensor(0.5887, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
68
Accuracy:
tensor(0.8375, dtype=torch.float64)
Loss:
tensor(0.5282, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
69
Accuracy:
tensor(0.8333, dtype=torch.float64)
Loss:
tensor(0.4947, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
70
Accuracy:
tensor(0.8333, dtype=torch.float64)
Loss:
tensor(0.5788, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
71
Accuracy:
tensor(0.8750, dtype=torch.float64)
Loss:
tensor(0.4855, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
72
Accuracy:
tensor(0.8292, dtype=torch.float64)
Loss:
tensor(0.4889, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
73
Accuracy:
tensor(0.7875, dtype=torch.float64)
Loss:
tensor(0.6304, grad_fn=<NllLossBackward>)
Epoch:
13
Batch:
74
Accuracy:
tensor(0.9417, dtype=torch.float64)
Loss:
tensor(0.3370, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
0
########################
Validation Accuracy:
tensor(0.2933, dtype=torch.float64)
########################
Accuracy:
tensor(0.9250, dtype=torch.float64)
Loss:
tensor(0.3228, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
1
Accuracy:
tensor(0.8583, dtype=torch.float64)
Loss:
tensor(0.4345, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
2
Accuracy:
tensor(0.9167, dtype=torch.float64)
Loss:
tensor(0.3631, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
3
Accuracy:
tensor(0.8917, dtype=torch.float64)
Loss:
tensor(0.3719, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
4
Accuracy:
tensor(0.8792, dtype=torch.float64)
Loss:
tensor(0.4112, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
5
Accuracy:
tensor(0.9042, dtype=torch.float64)
Loss:
tensor(0.3766, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
6
Accuracy:
tensor(0.9042, dtype=torch.float64)
Loss:
tensor(0.3680, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
7
Accuracy:
tensor(0.9000, dtype=torch.float64)
Loss:
tensor(0.3727, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
8
Accuracy:
tensor(0.9083, dtype=torch.float64)
Loss:
tensor(0.3910, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
9
Accuracy:
tensor(0.9333, dtype=torch.float64)
Loss:
tensor(0.3117, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
10
Accuracy:
tensor(0.9083, dtype=torch.float64)
Loss:
tensor(0.3570, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
11
Accuracy:
tensor(0.8542, dtype=torch.float64)
Loss:
tensor(0.4413, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
12
Accuracy:
tensor(0.9125, dtype=torch.float64)
Loss:
tensor(0.3491, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
13
Accuracy:
tensor(0.8750, dtype=torch.float64)
Loss:
tensor(0.4182, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
14
Accuracy:
tensor(0.8750, dtype=torch.float64)
Loss:
tensor(0.4168, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
15
Accuracy:
tensor(0.9250, dtype=torch.float64)
Loss:
tensor(0.3356, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
16
Accuracy:
tensor(0.9125, dtype=torch.float64)
Loss:
tensor(0.3946, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
17
Accuracy:
tensor(0.9042, dtype=torch.float64)
Loss:
tensor(0.3819, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
18
Accuracy:
tensor(0.9250, dtype=torch.float64)
Loss:
tensor(0.3342, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
19
Accuracy:
tensor(0.9042, dtype=torch.float64)
Loss:
tensor(0.3682, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
20
########################
Validation Accuracy:
tensor(0.3017, dtype=torch.float64)
########################
Accuracy:
tensor(0.9083, dtype=torch.float64)
Loss:
tensor(0.3585, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
21
Accuracy:
tensor(0.8958, dtype=torch.float64)
Loss:
tensor(0.3727, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
22
Accuracy:
tensor(0.8917, dtype=torch.float64)
Loss:
tensor(0.3734, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
23
Accuracy:
tensor(0.8958, dtype=torch.float64)
Loss:
tensor(0.3753, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
24
Accuracy:
tensor(0.8917, dtype=torch.float64)
Loss:
tensor(0.3923, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
25
Accuracy:
tensor(0.8625, dtype=torch.float64)
Loss:
tensor(0.4814, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
26
Accuracy:
tensor(0.9083, dtype=torch.float64)
Loss:
tensor(0.3698, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
27
Accuracy:
tensor(0.8792, dtype=torch.float64)
Loss:
tensor(0.4336, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
28
Accuracy:
tensor(0.8708, dtype=torch.float64)
Loss:
tensor(0.4636, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
29
Accuracy:
tensor(0.8875, dtype=torch.float64)
Loss:
tensor(0.4357, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
30
Accuracy:
tensor(0.9000, dtype=torch.float64)
Loss:
tensor(0.3740, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
31
Accuracy:
tensor(0.9083, dtype=torch.float64)
Loss:
tensor(0.3702, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
32
Accuracy:
tensor(0.8792, dtype=torch.float64)
Loss:
tensor(0.3957, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
33
Accuracy:
tensor(0.8833, dtype=torch.float64)
Loss:
tensor(0.4031, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
34
Accuracy:
tensor(0.9000, dtype=torch.float64)
Loss:
tensor(0.3856, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
35
Accuracy:
tensor(0.9000, dtype=torch.float64)
Loss:
tensor(0.3430, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
36
Accuracy:
tensor(0.8792, dtype=torch.float64)
Loss:
tensor(0.3895, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
37
Accuracy:
tensor(0.8708, dtype=torch.float64)
Loss:
tensor(0.4054, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
38
Accuracy:
tensor(0.8792, dtype=torch.float64)
Loss:
tensor(0.3783, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
39
Accuracy:
tensor(0.8667, dtype=torch.float64)
Loss:
tensor(0.4539, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
40
########################
Validation Accuracy:
tensor(0.2900, dtype=torch.float64)
########################
Accuracy:
tensor(0.8750, dtype=torch.float64)
Loss:
tensor(0.4014, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
41
Accuracy:
tensor(0.8917, dtype=torch.float64)
Loss:
tensor(0.4252, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
42
Accuracy:
tensor(0.8875, dtype=torch.float64)
Loss:
tensor(0.3881, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
43
Accuracy:
tensor(0.8875, dtype=torch.float64)
Loss:
tensor(0.3883, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
44
Accuracy:
tensor(0.9083, dtype=torch.float64)
Loss:
tensor(0.3784, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
45
Accuracy:
tensor(0.8583, dtype=torch.float64)
Loss:
tensor(0.4404, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
46
Accuracy:
tensor(0.8875, dtype=torch.float64)
Loss:
tensor(0.3993, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
47
Accuracy:
tensor(0.8792, dtype=torch.float64)
Loss:
tensor(0.4074, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
48
Accuracy:
tensor(0.9000, dtype=torch.float64)
Loss:
tensor(0.3480, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
49
Accuracy:
tensor(0.8500, dtype=torch.float64)
Loss:
tensor(0.4521, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
50
Accuracy:
tensor(0.8625, dtype=torch.float64)
Loss:
tensor(0.4405, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
51
Accuracy:
tensor(0.8708, dtype=torch.float64)
Loss:
tensor(0.4378, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
52
Accuracy:
tensor(0.8958, dtype=torch.float64)
Loss:
tensor(0.4582, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
53
Accuracy:
tensor(0.8667, dtype=torch.float64)
Loss:
tensor(0.4174, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
54
Accuracy:
tensor(0.8208, dtype=torch.float64)
Loss:
tensor(0.5090, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
55
Accuracy:
tensor(0.8875, dtype=torch.float64)
Loss:
tensor(0.4335, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
56
Accuracy:
tensor(0.8875, dtype=torch.float64)
Loss:
tensor(0.4489, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
57
Accuracy:
tensor(0.8750, dtype=torch.float64)
Loss:
tensor(0.4381, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
58
Accuracy:
tensor(0.8792, dtype=torch.float64)
Loss:
tensor(0.4320, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
59
Accuracy:
tensor(0.8958, dtype=torch.float64)
Loss:
tensor(0.4145, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
60
########################
Validation Accuracy:
tensor(0.2917, dtype=torch.float64)
########################
Accuracy:
tensor(0.8750, dtype=torch.float64)
Loss:
tensor(0.4038, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
61
Accuracy:
tensor(0.9000, dtype=torch.float64)
Loss:
tensor(0.4072, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
62
Accuracy:
tensor(0.8583, dtype=torch.float64)
Loss:
tensor(0.4830, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
63
Accuracy:
tensor(0.8208, dtype=torch.float64)
Loss:
tensor(0.5156, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
64
Accuracy:
tensor(0.8583, dtype=torch.float64)
Loss:
tensor(0.4468, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
65
Accuracy:
tensor(0.8292, dtype=torch.float64)
Loss:
tensor(0.5042, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
66
Accuracy:
tensor(0.8917, dtype=torch.float64)
Loss:
tensor(0.3776, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
67
Accuracy:
tensor(0.8958, dtype=torch.float64)
Loss:
tensor(0.3873, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
68
Accuracy:
tensor(0.8542, dtype=torch.float64)
Loss:
tensor(0.4944, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
69
Accuracy:
tensor(0.8958, dtype=torch.float64)
Loss:
tensor(0.3604, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
70
Accuracy:
tensor(0.8500, dtype=torch.float64)
Loss:
tensor(0.4822, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
71
Accuracy:
tensor(0.8333, dtype=torch.float64)
Loss:
tensor(0.5179, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
72
Accuracy:
tensor(0.8583, dtype=torch.float64)
Loss:
tensor(0.4737, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
73
Accuracy:
tensor(0.8208, dtype=torch.float64)
Loss:
tensor(0.5517, grad_fn=<NllLossBackward>)
Epoch:
14
Batch:
74
Accuracy:
tensor(0.8958, dtype=torch.float64)
Loss:
tensor(0.3728, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
0
########################
Validation Accuracy:
tensor(0.3050, dtype=torch.float64)
########################
Accuracy:
tensor(0.9292, dtype=torch.float64)
Loss:
tensor(0.3014, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
1
Accuracy:
tensor(0.9083, dtype=torch.float64)
Loss:
tensor(0.3389, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
2
Accuracy:
tensor(0.9417, dtype=torch.float64)
Loss:
tensor(0.2685, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
3
Accuracy:
tensor(0.9167, dtype=torch.float64)
Loss:
tensor(0.2982, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
4
Accuracy:
tensor(0.8958, dtype=torch.float64)
Loss:
tensor(0.3521, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
5
Accuracy:
tensor(0.9417, dtype=torch.float64)
Loss:
tensor(0.2804, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
6
Accuracy:
tensor(0.9000, dtype=torch.float64)
Loss:
tensor(0.3453, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
7
Accuracy:
tensor(0.9208, dtype=torch.float64)
Loss:
tensor(0.3052, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
8
Accuracy:
tensor(0.9333, dtype=torch.float64)
Loss:
tensor(0.2945, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
9
Accuracy:
tensor(0.9375, dtype=torch.float64)
Loss:
tensor(0.3042, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
10
Accuracy:
tensor(0.9167, dtype=torch.float64)
Loss:
tensor(0.3111, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
11
Accuracy:
tensor(0.9417, dtype=torch.float64)
Loss:
tensor(0.3201, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
12
Accuracy:
tensor(0.9333, dtype=torch.float64)
Loss:
tensor(0.2978, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
13
Accuracy:
tensor(0.9125, dtype=torch.float64)
Loss:
tensor(0.3536, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
14
Accuracy:
tensor(0.9083, dtype=torch.float64)
Loss:
tensor(0.3482, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
15
Accuracy:
tensor(0.9250, dtype=torch.float64)
Loss:
tensor(0.3002, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
16
Accuracy:
tensor(0.8958, dtype=torch.float64)
Loss:
tensor(0.3601, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
17
Accuracy:
tensor(0.9250, dtype=torch.float64)
Loss:
tensor(0.3612, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
18
Accuracy:
tensor(0.9083, dtype=torch.float64)
Loss:
tensor(0.3244, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
19
Accuracy:
tensor(0.9083, dtype=torch.float64)
Loss:
tensor(0.3255, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
20
########################
Validation Accuracy:
tensor(0.3167, dtype=torch.float64)
########################
Accuracy:
tensor(0.9125, dtype=torch.float64)
Loss:
tensor(0.3221, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
21
Accuracy:
tensor(0.9125, dtype=torch.float64)
Loss:
tensor(0.3782, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
22
Accuracy:
tensor(0.9292, dtype=torch.float64)
Loss:
tensor(0.3415, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
23
Accuracy:
tensor(0.9125, dtype=torch.float64)
Loss:
tensor(0.3458, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
24
Accuracy:
tensor(0.9292, dtype=torch.float64)
Loss:
tensor(0.3133, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
25
Accuracy:
tensor(0.8792, dtype=torch.float64)
Loss:
tensor(0.3523, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
26
Accuracy:
tensor(0.9000, dtype=torch.float64)
Loss:
tensor(0.3701, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
27
Accuracy:
tensor(0.8917, dtype=torch.float64)
Loss:
tensor(0.3389, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
28
Accuracy:
tensor(0.9083, dtype=torch.float64)
Loss:
tensor(0.3425, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
29
Accuracy:
tensor(0.8792, dtype=torch.float64)
Loss:
tensor(0.3890, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
30
Accuracy:
tensor(0.9208, dtype=torch.float64)
Loss:
tensor(0.2941, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
31
Accuracy:
tensor(0.8750, dtype=torch.float64)
Loss:
tensor(0.3751, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
32
Accuracy:
tensor(0.9042, dtype=torch.float64)
Loss:
tensor(0.2953, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
33
Accuracy:
tensor(0.9250, dtype=torch.float64)
Loss:
tensor(0.3232, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
34
Accuracy:
tensor(0.9083, dtype=torch.float64)
Loss:
tensor(0.3394, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
35
Accuracy:
tensor(0.9208, dtype=torch.float64)
Loss:
tensor(0.2940, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
36
Accuracy:
tensor(0.9000, dtype=torch.float64)
Loss:
tensor(0.3457, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
37
Accuracy:
tensor(0.9000, dtype=torch.float64)
Loss:
tensor(0.3435, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
38
Accuracy:
tensor(0.8500, dtype=torch.float64)
Loss:
tensor(0.4313, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
39
Accuracy:
tensor(0.8417, dtype=torch.float64)
Loss:
tensor(0.4379, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
40
########################
Validation Accuracy:
tensor(0.3050, dtype=torch.float64)
########################
Accuracy:
tensor(0.9000, dtype=torch.float64)
Loss:
tensor(0.3211, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
41
Accuracy:
tensor(0.8792, dtype=torch.float64)
Loss:
tensor(0.4092, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
42
Accuracy:
tensor(0.8792, dtype=torch.float64)
Loss:
tensor(0.4185, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
43
Accuracy:
tensor(0.9208, dtype=torch.float64)
Loss:
tensor(0.3199, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
44
Accuracy:
tensor(0.8833, dtype=torch.float64)
Loss:
tensor(0.4009, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
45
Accuracy:
tensor(0.9042, dtype=torch.float64)
Loss:
tensor(0.3495, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
46
Accuracy:
tensor(0.8958, dtype=torch.float64)
Loss:
tensor(0.3229, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
47
Accuracy:
tensor(0.8708, dtype=torch.float64)
Loss:
tensor(0.3952, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
48
Accuracy:
tensor(0.8875, dtype=torch.float64)
Loss:
tensor(0.3838, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
49
Accuracy:
tensor(0.9250, dtype=torch.float64)
Loss:
tensor(0.3176, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
50
Accuracy:
tensor(0.8833, dtype=torch.float64)
Loss:
tensor(0.3809, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
51
Accuracy:
tensor(0.8833, dtype=torch.float64)
Loss:
tensor(0.4080, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
52
Accuracy:
tensor(0.8958, dtype=torch.float64)
Loss:
tensor(0.3415, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
53
Accuracy:
tensor(0.8792, dtype=torch.float64)
Loss:
tensor(0.3493, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
54
Accuracy:
tensor(0.8833, dtype=torch.float64)
Loss:
tensor(0.3522, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
55
Accuracy:
tensor(0.9000, dtype=torch.float64)
Loss:
tensor(0.3223, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
56
Accuracy:
tensor(0.8583, dtype=torch.float64)
Loss:
tensor(0.4255, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
57
Accuracy:
tensor(0.8792, dtype=torch.float64)
Loss:
tensor(0.4010, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
58
Accuracy:
tensor(0.8875, dtype=torch.float64)
Loss:
tensor(0.4085, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
59
Accuracy:
tensor(0.8792, dtype=torch.float64)
Loss:
tensor(0.3879, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
60
########################
Validation Accuracy:
tensor(0.2917, dtype=torch.float64)
########################
Accuracy:
tensor(0.8417, dtype=torch.float64)
Loss:
tensor(0.4794, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
61
Accuracy:
tensor(0.8583, dtype=torch.float64)
Loss:
tensor(0.4173, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
62
Accuracy:
tensor(0.9000, dtype=torch.float64)
Loss:
tensor(0.4365, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
63
Accuracy:
tensor(0.8625, dtype=torch.float64)
Loss:
tensor(0.4399, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
64
Accuracy:
tensor(0.8417, dtype=torch.float64)
Loss:
tensor(0.4660, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
65
Accuracy:
tensor(0.8708, dtype=torch.float64)
Loss:
tensor(0.4240, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
66
Accuracy:
tensor(0.8542, dtype=torch.float64)
Loss:
tensor(0.4228, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
67
Accuracy:
tensor(0.8958, dtype=torch.float64)
Loss:
tensor(0.3439, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
68
Accuracy:
tensor(0.8625, dtype=torch.float64)
Loss:
tensor(0.4300, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
69
Accuracy:
tensor(0.9083, dtype=torch.float64)
Loss:
tensor(0.4320, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
70
Accuracy:
tensor(0.8500, dtype=torch.float64)
Loss:
tensor(0.4651, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
71
Accuracy:
tensor(0.8667, dtype=torch.float64)
Loss:
tensor(0.4032, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
72
Accuracy:
tensor(0.9042, dtype=torch.float64)
Loss:
tensor(0.3693, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
73
Accuracy:
tensor(0.8875, dtype=torch.float64)
Loss:
tensor(0.3661, grad_fn=<NllLossBackward>)
Epoch:
15
Batch:
74
Accuracy:
tensor(0.9333, dtype=torch.float64)
Loss:
tensor(0.2547, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
0
########################
Validation Accuracy:
tensor(0.3000, dtype=torch.float64)
########################
Accuracy:
tensor(0.9292, dtype=torch.float64)
Loss:
tensor(0.2840, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
1
Accuracy:
tensor(0.9417, dtype=torch.float64)
Loss:
tensor(0.2324, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
2
Accuracy:
tensor(0.9167, dtype=torch.float64)
Loss:
tensor(0.2838, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
3
Accuracy:
tensor(0.9333, dtype=torch.float64)
Loss:
tensor(0.2691, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
4
Accuracy:
tensor(0.9458, dtype=torch.float64)
Loss:
tensor(0.2652, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
5
Accuracy:
tensor(0.9292, dtype=torch.float64)
Loss:
tensor(0.2843, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
6
Accuracy:
tensor(0.9333, dtype=torch.float64)
Loss:
tensor(0.2335, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
7
Accuracy:
tensor(0.9333, dtype=torch.float64)
Loss:
tensor(0.2708, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
8
Accuracy:
tensor(0.9292, dtype=torch.float64)
Loss:
tensor(0.2393, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
9
Accuracy:
tensor(0.9375, dtype=torch.float64)
Loss:
tensor(0.2551, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
10
Accuracy:
tensor(0.9000, dtype=torch.float64)
Loss:
tensor(0.3008, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
11
Accuracy:
tensor(0.9333, dtype=torch.float64)
Loss:
tensor(0.2807, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
12
Accuracy:
tensor(0.9250, dtype=torch.float64)
Loss:
tensor(0.2821, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
13
Accuracy:
tensor(0.9417, dtype=torch.float64)
Loss:
tensor(0.2562, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
14
Accuracy:
tensor(0.9083, dtype=torch.float64)
Loss:
tensor(0.3188, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
15
Accuracy:
tensor(0.9375, dtype=torch.float64)
Loss:
tensor(0.2450, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
16
Accuracy:
tensor(0.9417, dtype=torch.float64)
Loss:
tensor(0.2699, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
17
Accuracy:
tensor(0.9250, dtype=torch.float64)
Loss:
tensor(0.2913, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
18
Accuracy:
tensor(0.9292, dtype=torch.float64)
Loss:
tensor(0.2712, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
19
Accuracy:
tensor(0.9083, dtype=torch.float64)
Loss:
tensor(0.2710, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
20
########################
Validation Accuracy:
tensor(0.3083, dtype=torch.float64)
########################
Accuracy:
tensor(0.9167, dtype=torch.float64)
Loss:
tensor(0.3193, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
21
Accuracy:
tensor(0.9292, dtype=torch.float64)
Loss:
tensor(0.2918, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
22
Accuracy:
tensor(0.9125, dtype=torch.float64)
Loss:
tensor(0.2787, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
23
Accuracy:
tensor(0.9292, dtype=torch.float64)
Loss:
tensor(0.2798, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
24
Accuracy:
tensor(0.9583, dtype=torch.float64)
Loss:
tensor(0.2239, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
25
Accuracy:
tensor(0.9458, dtype=torch.float64)
Loss:
tensor(0.2547, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
26
Accuracy:
tensor(0.9375, dtype=torch.float64)
Loss:
tensor(0.2564, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
27
Accuracy:
tensor(0.9167, dtype=torch.float64)
Loss:
tensor(0.2584, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
28
Accuracy:
tensor(0.9042, dtype=torch.float64)
Loss:
tensor(0.3472, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
29
Accuracy:
tensor(0.9333, dtype=torch.float64)
Loss:
tensor(0.2628, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
30
Accuracy:
tensor(0.9083, dtype=torch.float64)
Loss:
tensor(0.3100, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
31
Accuracy:
tensor(0.9292, dtype=torch.float64)
Loss:
tensor(0.2684, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
32
Accuracy:
tensor(0.9167, dtype=torch.float64)
Loss:
tensor(0.3008, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
33
Accuracy:
tensor(0.9208, dtype=torch.float64)
Loss:
tensor(0.2911, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
34
Accuracy:
tensor(0.9208, dtype=torch.float64)
Loss:
tensor(0.2691, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
35
Accuracy:
tensor(0.9208, dtype=torch.float64)
Loss:
tensor(0.2766, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
36
Accuracy:
tensor(0.9042, dtype=torch.float64)
Loss:
tensor(0.2793, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
37
Accuracy:
tensor(0.9375, dtype=torch.float64)
Loss:
tensor(0.2717, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
38
Accuracy:
tensor(0.9417, dtype=torch.float64)
Loss:
tensor(0.2589, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
39
Accuracy:
tensor(0.9292, dtype=torch.float64)
Loss:
tensor(0.2487, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
40
########################
Validation Accuracy:
tensor(0.2900, dtype=torch.float64)
########################
Accuracy:
tensor(0.9292, dtype=torch.float64)
Loss:
tensor(0.3017, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
41
Accuracy:
tensor(0.9000, dtype=torch.float64)
Loss:
tensor(0.3452, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
42
Accuracy:
tensor(0.8958, dtype=torch.float64)
Loss:
tensor(0.2964, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
43
Accuracy:
tensor(0.9167, dtype=torch.float64)
Loss:
tensor(0.2995, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
44
Accuracy:
tensor(0.9250, dtype=torch.float64)
Loss:
tensor(0.2991, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
45
Accuracy:
tensor(0.9042, dtype=torch.float64)
Loss:
tensor(0.3098, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
46
Accuracy:
tensor(0.9083, dtype=torch.float64)
Loss:
tensor(0.3413, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
47
Accuracy:
tensor(0.8583, dtype=torch.float64)
Loss:
tensor(0.4267, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
48
Accuracy:
tensor(0.8917, dtype=torch.float64)
Loss:
tensor(0.3572, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
49
Accuracy:
tensor(0.9167, dtype=torch.float64)
Loss:
tensor(0.2985, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
50
Accuracy:
tensor(0.9292, dtype=torch.float64)
Loss:
tensor(0.2821, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
51
Accuracy:
tensor(0.9042, dtype=torch.float64)
Loss:
tensor(0.3664, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
52
Accuracy:
tensor(0.8875, dtype=torch.float64)
Loss:
tensor(0.3761, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
53
Accuracy:
tensor(0.8958, dtype=torch.float64)
Loss:
tensor(0.3690, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
54
Accuracy:
tensor(0.8875, dtype=torch.float64)
Loss:
tensor(0.3918, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
55
Accuracy:
tensor(0.9167, dtype=torch.float64)
Loss:
tensor(0.3211, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
56
Accuracy:
tensor(0.9292, dtype=torch.float64)
Loss:
tensor(0.2813, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
57
Accuracy:
tensor(0.8708, dtype=torch.float64)
Loss:
tensor(0.3830, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
58
Accuracy:
tensor(0.9083, dtype=torch.float64)
Loss:
tensor(0.3172, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
59
Accuracy:
tensor(0.8833, dtype=torch.float64)
Loss:
tensor(0.3438, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
60
########################
Validation Accuracy:
tensor(0.2967, dtype=torch.float64)
########################
Accuracy:
tensor(0.8792, dtype=torch.float64)
Loss:
tensor(0.3715, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
61
Accuracy:
tensor(0.9208, dtype=torch.float64)
Loss:
tensor(0.3415, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
62
Accuracy:
tensor(0.9042, dtype=torch.float64)
Loss:
tensor(0.3097, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
63
Accuracy:
tensor(0.8708, dtype=torch.float64)
Loss:
tensor(0.3828, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
64
Accuracy:
tensor(0.9042, dtype=torch.float64)
Loss:
tensor(0.3437, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
65
Accuracy:
tensor(0.8667, dtype=torch.float64)
Loss:
tensor(0.3588, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
66
Accuracy:
tensor(0.9042, dtype=torch.float64)
Loss:
tensor(0.3493, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
67
Accuracy:
tensor(0.9250, dtype=torch.float64)
Loss:
tensor(0.2781, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
68
Accuracy:
tensor(0.9000, dtype=torch.float64)
Loss:
tensor(0.3386, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
69
Accuracy:
tensor(0.8750, dtype=torch.float64)
Loss:
tensor(0.4007, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
70
Accuracy:
tensor(0.9083, dtype=torch.float64)
Loss:
tensor(0.2947, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
71
Accuracy:
tensor(0.8792, dtype=torch.float64)
Loss:
tensor(0.3781, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
72
Accuracy:
tensor(0.8875, dtype=torch.float64)
Loss:
tensor(0.3700, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
73
Accuracy:
tensor(0.9083, dtype=torch.float64)
Loss:
tensor(0.3216, grad_fn=<NllLossBackward>)
Epoch:
16
Batch:
74
Accuracy:
tensor(0.9250, dtype=torch.float64)
Loss:
tensor(0.2892, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
0
########################
Validation Accuracy:
tensor(0.3150, dtype=torch.float64)
########################
Accuracy:
tensor(0.9542, dtype=torch.float64)
Loss:
tensor(0.2094, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
1
Accuracy:
tensor(0.9458, dtype=torch.float64)
Loss:
tensor(0.2630, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
2
Accuracy:
tensor(0.9542, dtype=torch.float64)
Loss:
tensor(0.2085, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
3
Accuracy:
tensor(0.9625, dtype=torch.float64)
Loss:
tensor(0.2280, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
4
Accuracy:
tensor(0.9625, dtype=torch.float64)
Loss:
tensor(0.2061, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
5
Accuracy:
tensor(0.9542, dtype=torch.float64)
Loss:
tensor(0.1969, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
6
Accuracy:
tensor(0.9625, dtype=torch.float64)
Loss:
tensor(0.1990, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
7
Accuracy:
tensor(0.9500, dtype=torch.float64)
Loss:
tensor(0.2095, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
8
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.1945, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
9
Accuracy:
tensor(0.9458, dtype=torch.float64)
Loss:
tensor(0.2363, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
10
Accuracy:
tensor(0.9458, dtype=torch.float64)
Loss:
tensor(0.2464, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
11
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.1981, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
12
Accuracy:
tensor(0.9375, dtype=torch.float64)
Loss:
tensor(0.2896, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
13
Accuracy:
tensor(0.9417, dtype=torch.float64)
Loss:
tensor(0.2371, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
14
Accuracy:
tensor(0.9250, dtype=torch.float64)
Loss:
tensor(0.2776, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
15
Accuracy:
tensor(0.9250, dtype=torch.float64)
Loss:
tensor(0.2630, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
16
Accuracy:
tensor(0.9458, dtype=torch.float64)
Loss:
tensor(0.2372, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
17
Accuracy:
tensor(0.9167, dtype=torch.float64)
Loss:
tensor(0.2727, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
18
Accuracy:
tensor(0.9417, dtype=torch.float64)
Loss:
tensor(0.2514, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
19
Accuracy:
tensor(0.9500, dtype=torch.float64)
Loss:
tensor(0.2038, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
20
########################
Validation Accuracy:
tensor(0.2950, dtype=torch.float64)
########################
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.1883, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
21
Accuracy:
tensor(0.9375, dtype=torch.float64)
Loss:
tensor(0.2312, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
22
Accuracy:
tensor(0.9625, dtype=torch.float64)
Loss:
tensor(0.1952, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
23
Accuracy:
tensor(0.9167, dtype=torch.float64)
Loss:
tensor(0.2844, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
24
Accuracy:
tensor(0.9500, dtype=torch.float64)
Loss:
tensor(0.2170, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
25
Accuracy:
tensor(0.9042, dtype=torch.float64)
Loss:
tensor(0.3170, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
26
Accuracy:
tensor(0.9500, dtype=torch.float64)
Loss:
tensor(0.2639, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
27
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.1839, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
28
Accuracy:
tensor(0.9250, dtype=torch.float64)
Loss:
tensor(0.2662, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
29
Accuracy:
tensor(0.9458, dtype=torch.float64)
Loss:
tensor(0.2247, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
30
Accuracy:
tensor(0.9333, dtype=torch.float64)
Loss:
tensor(0.2668, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
31
Accuracy:
tensor(0.9500, dtype=torch.float64)
Loss:
tensor(0.2275, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
32
Accuracy:
tensor(0.9333, dtype=torch.float64)
Loss:
tensor(0.2902, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
33
Accuracy:
tensor(0.9042, dtype=torch.float64)
Loss:
tensor(0.2920, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
34
Accuracy:
tensor(0.9167, dtype=torch.float64)
Loss:
tensor(0.2765, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
35
Accuracy:
tensor(0.9375, dtype=torch.float64)
Loss:
tensor(0.2673, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
36
Accuracy:
tensor(0.9500, dtype=torch.float64)
Loss:
tensor(0.2634, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
37
Accuracy:
tensor(0.9333, dtype=torch.float64)
Loss:
tensor(0.2513, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
38
Accuracy:
tensor(0.9167, dtype=torch.float64)
Loss:
tensor(0.3097, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
39
Accuracy:
tensor(0.8833, dtype=torch.float64)
Loss:
tensor(0.3081, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
40
########################
Validation Accuracy:
tensor(0.3033, dtype=torch.float64)
########################
Accuracy:
tensor(0.8917, dtype=torch.float64)
Loss:
tensor(0.3231, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
41
Accuracy:
tensor(0.9333, dtype=torch.float64)
Loss:
tensor(0.2881, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
42
Accuracy:
tensor(0.8917, dtype=torch.float64)
Loss:
tensor(0.3162, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
43
Accuracy:
tensor(0.8958, dtype=torch.float64)
Loss:
tensor(0.3294, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
44
Accuracy:
tensor(0.9542, dtype=torch.float64)
Loss:
tensor(0.2215, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
45
Accuracy:
tensor(0.8958, dtype=torch.float64)
Loss:
tensor(0.3081, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
46
Accuracy:
tensor(0.9167, dtype=torch.float64)
Loss:
tensor(0.2885, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
47
Accuracy:
tensor(0.9208, dtype=torch.float64)
Loss:
tensor(0.2844, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
48
Accuracy:
tensor(0.9208, dtype=torch.float64)
Loss:
tensor(0.2716, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
49
Accuracy:
tensor(0.9292, dtype=torch.float64)
Loss:
tensor(0.2866, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
50
Accuracy:
tensor(0.9125, dtype=torch.float64)
Loss:
tensor(0.3042, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
51
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.2230, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
52
Accuracy:
tensor(0.9167, dtype=torch.float64)
Loss:
tensor(0.2827, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
53
Accuracy:
tensor(0.9167, dtype=torch.float64)
Loss:
tensor(0.3050, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
54
Accuracy:
tensor(0.9333, dtype=torch.float64)
Loss:
tensor(0.2586, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
55
Accuracy:
tensor(0.9208, dtype=torch.float64)
Loss:
tensor(0.2610, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
56
Accuracy:
tensor(0.9375, dtype=torch.float64)
Loss:
tensor(0.3083, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
57
Accuracy:
tensor(0.9375, dtype=torch.float64)
Loss:
tensor(0.2712, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
58
Accuracy:
tensor(0.9125, dtype=torch.float64)
Loss:
tensor(0.3051, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
59
Accuracy:
tensor(0.9292, dtype=torch.float64)
Loss:
tensor(0.2716, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
60
########################
Validation Accuracy:
tensor(0.3017, dtype=torch.float64)
########################
Accuracy:
tensor(0.8875, dtype=torch.float64)
Loss:
tensor(0.3364, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
61
Accuracy:
tensor(0.8875, dtype=torch.float64)
Loss:
tensor(0.3242, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
62
Accuracy:
tensor(0.9000, dtype=torch.float64)
Loss:
tensor(0.3439, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
63
Accuracy:
tensor(0.9167, dtype=torch.float64)
Loss:
tensor(0.2957, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
64
Accuracy:
tensor(0.9167, dtype=torch.float64)
Loss:
tensor(0.2609, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
65
Accuracy:
tensor(0.9167, dtype=torch.float64)
Loss:
tensor(0.2638, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
66
Accuracy:
tensor(0.8750, dtype=torch.float64)
Loss:
tensor(0.3002, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
67
Accuracy:
tensor(0.9250, dtype=torch.float64)
Loss:
tensor(0.3225, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
68
Accuracy:
tensor(0.9000, dtype=torch.float64)
Loss:
tensor(0.3627, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
69
Accuracy:
tensor(0.8750, dtype=torch.float64)
Loss:
tensor(0.3416, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
70
Accuracy:
tensor(0.9250, dtype=torch.float64)
Loss:
tensor(0.3385, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
71
Accuracy:
tensor(0.9000, dtype=torch.float64)
Loss:
tensor(0.3478, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
72
Accuracy:
tensor(0.9250, dtype=torch.float64)
Loss:
tensor(0.2831, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
73
Accuracy:
tensor(0.9042, dtype=torch.float64)
Loss:
tensor(0.3151, grad_fn=<NllLossBackward>)
Epoch:
17
Batch:
74
Accuracy:
tensor(0.9417, dtype=torch.float64)
Loss:
tensor(0.2054, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
0
########################
Validation Accuracy:
tensor(0.3000, dtype=torch.float64)
########################
Accuracy:
tensor(0.9417, dtype=torch.float64)
Loss:
tensor(0.2406, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
1
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.1850, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
2
Accuracy:
tensor(0.9333, dtype=torch.float64)
Loss:
tensor(0.2475, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
3
Accuracy:
tensor(0.9542, dtype=torch.float64)
Loss:
tensor(0.1851, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
4
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.2026, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
5
Accuracy:
tensor(0.9250, dtype=torch.float64)
Loss:
tensor(0.2423, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
6
Accuracy:
tensor(0.9542, dtype=torch.float64)
Loss:
tensor(0.1722, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
7
Accuracy:
tensor(0.9500, dtype=torch.float64)
Loss:
tensor(0.1821, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
8
Accuracy:
tensor(0.9458, dtype=torch.float64)
Loss:
tensor(0.2624, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
9
Accuracy:
tensor(0.9625, dtype=torch.float64)
Loss:
tensor(0.1878, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
10
Accuracy:
tensor(0.8833, dtype=torch.float64)
Loss:
tensor(0.3194, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
11
Accuracy:
tensor(0.8958, dtype=torch.float64)
Loss:
tensor(0.3082, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
12
Accuracy:
tensor(0.9542, dtype=torch.float64)
Loss:
tensor(0.1876, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
13
Accuracy:
tensor(0.9208, dtype=torch.float64)
Loss:
tensor(0.2900, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
14
Accuracy:
tensor(0.9625, dtype=torch.float64)
Loss:
tensor(0.2018, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
15
Accuracy:
tensor(0.9542, dtype=torch.float64)
Loss:
tensor(0.1844, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
16
Accuracy:
tensor(0.9458, dtype=torch.float64)
Loss:
tensor(0.2188, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
17
Accuracy:
tensor(0.9167, dtype=torch.float64)
Loss:
tensor(0.2643, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
18
Accuracy:
tensor(0.9625, dtype=torch.float64)
Loss:
tensor(0.1984, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
19
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.2005, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
20
########################
Validation Accuracy:
tensor(0.3017, dtype=torch.float64)
########################
Accuracy:
tensor(0.9333, dtype=torch.float64)
Loss:
tensor(0.2626, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
21
Accuracy:
tensor(0.9542, dtype=torch.float64)
Loss:
tensor(0.2162, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
22
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.1738, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
23
Accuracy:
tensor(0.9458, dtype=torch.float64)
Loss:
tensor(0.1950, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
24
Accuracy:
tensor(0.9292, dtype=torch.float64)
Loss:
tensor(0.2294, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
25
Accuracy:
tensor(0.9292, dtype=torch.float64)
Loss:
tensor(0.2555, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
26
Accuracy:
tensor(0.9583, dtype=torch.float64)
Loss:
tensor(0.1862, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
27
Accuracy:
tensor(0.9333, dtype=torch.float64)
Loss:
tensor(0.2630, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
28
Accuracy:
tensor(0.9333, dtype=torch.float64)
Loss:
tensor(0.2908, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
29
Accuracy:
tensor(0.9500, dtype=torch.float64)
Loss:
tensor(0.2034, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
30
Accuracy:
tensor(0.9458, dtype=torch.float64)
Loss:
tensor(0.2051, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
31
Accuracy:
tensor(0.8958, dtype=torch.float64)
Loss:
tensor(0.3146, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
32
Accuracy:
tensor(0.9375, dtype=torch.float64)
Loss:
tensor(0.2727, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
33
Accuracy:
tensor(0.9167, dtype=torch.float64)
Loss:
tensor(0.2681, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
34
Accuracy:
tensor(0.9042, dtype=torch.float64)
Loss:
tensor(0.2843, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
35
Accuracy:
tensor(0.9583, dtype=torch.float64)
Loss:
tensor(0.1973, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
36
Accuracy:
tensor(0.9500, dtype=torch.float64)
Loss:
tensor(0.2091, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
37
Accuracy:
tensor(0.9000, dtype=torch.float64)
Loss:
tensor(0.3027, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
38
Accuracy:
tensor(0.9250, dtype=torch.float64)
Loss:
tensor(0.2598, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
39
Accuracy:
tensor(0.9292, dtype=torch.float64)
Loss:
tensor(0.2384, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
40
########################
Validation Accuracy:
tensor(0.2950, dtype=torch.float64)
########################
Accuracy:
tensor(0.9167, dtype=torch.float64)
Loss:
tensor(0.2633, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
41
Accuracy:
tensor(0.9250, dtype=torch.float64)
Loss:
tensor(0.2646, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
42
Accuracy:
tensor(0.9500, dtype=torch.float64)
Loss:
tensor(0.2214, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
43
Accuracy:
tensor(0.9500, dtype=torch.float64)
Loss:
tensor(0.2176, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
44
Accuracy:
tensor(0.9167, dtype=torch.float64)
Loss:
tensor(0.2885, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
45
Accuracy:
tensor(0.9167, dtype=torch.float64)
Loss:
tensor(0.2715, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
46
Accuracy:
tensor(0.9417, dtype=torch.float64)
Loss:
tensor(0.2205, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
47
Accuracy:
tensor(0.9500, dtype=torch.float64)
Loss:
tensor(0.2150, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
48
Accuracy:
tensor(0.9125, dtype=torch.float64)
Loss:
tensor(0.3027, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
49
Accuracy:
tensor(0.9125, dtype=torch.float64)
Loss:
tensor(0.2465, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
50
Accuracy:
tensor(0.9542, dtype=torch.float64)
Loss:
tensor(0.2321, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
51
Accuracy:
tensor(0.9125, dtype=torch.float64)
Loss:
tensor(0.2708, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
52
Accuracy:
tensor(0.9458, dtype=torch.float64)
Loss:
tensor(0.2258, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
53
Accuracy:
tensor(0.9000, dtype=torch.float64)
Loss:
tensor(0.2788, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
54
Accuracy:
tensor(0.9542, dtype=torch.float64)
Loss:
tensor(0.2447, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
55
Accuracy:
tensor(0.9333, dtype=torch.float64)
Loss:
tensor(0.2862, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
56
Accuracy:
tensor(0.9417, dtype=torch.float64)
Loss:
tensor(0.2320, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
57
Accuracy:
tensor(0.8958, dtype=torch.float64)
Loss:
tensor(0.3034, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
58
Accuracy:
tensor(0.9208, dtype=torch.float64)
Loss:
tensor(0.2508, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
59
Accuracy:
tensor(0.9250, dtype=torch.float64)
Loss:
tensor(0.2498, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
60
########################
Validation Accuracy:
tensor(0.2967, dtype=torch.float64)
########################
Accuracy:
tensor(0.9208, dtype=torch.float64)
Loss:
tensor(0.2619, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
61
Accuracy:
tensor(0.9000, dtype=torch.float64)
Loss:
tensor(0.3022, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
62
Accuracy:
tensor(0.9417, dtype=torch.float64)
Loss:
tensor(0.2139, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
63
Accuracy:
tensor(0.9167, dtype=torch.float64)
Loss:
tensor(0.2809, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
64
Accuracy:
tensor(0.8708, dtype=torch.float64)
Loss:
tensor(0.3425, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
65
Accuracy:
tensor(0.9208, dtype=torch.float64)
Loss:
tensor(0.2916, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
66
Accuracy:
tensor(0.9417, dtype=torch.float64)
Loss:
tensor(0.2236, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
67
Accuracy:
tensor(0.8708, dtype=torch.float64)
Loss:
tensor(0.3410, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
68
Accuracy:
tensor(0.9500, dtype=torch.float64)
Loss:
tensor(0.2090, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
69
Accuracy:
tensor(0.9250, dtype=torch.float64)
Loss:
tensor(0.2325, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
70
Accuracy:
tensor(0.8833, dtype=torch.float64)
Loss:
tensor(0.2956, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
71
Accuracy:
tensor(0.9125, dtype=torch.float64)
Loss:
tensor(0.3515, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
72
Accuracy:
tensor(0.9250, dtype=torch.float64)
Loss:
tensor(0.2863, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
73
Accuracy:
tensor(0.9417, dtype=torch.float64)
Loss:
tensor(0.2312, grad_fn=<NllLossBackward>)
Epoch:
18
Batch:
74
Accuracy:
tensor(0.9250, dtype=torch.float64)
Loss:
tensor(0.2520, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
0
########################
Validation Accuracy:
tensor(0.2967, dtype=torch.float64)
########################
Accuracy:
tensor(0.9625, dtype=torch.float64)
Loss:
tensor(0.1772, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
1
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.1593, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
2
Accuracy:
tensor(0.9625, dtype=torch.float64)
Loss:
tensor(0.1600, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
3
Accuracy:
tensor(0.9375, dtype=torch.float64)
Loss:
tensor(0.2132, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
4
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.1758, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
5
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.1419, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
6
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.1575, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
7
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.1987, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
8
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.1568, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
9
Accuracy:
tensor(0.9583, dtype=torch.float64)
Loss:
tensor(0.1952, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
10
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.1675, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
11
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.1449, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
12
Accuracy:
tensor(0.9625, dtype=torch.float64)
Loss:
tensor(0.1959, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
13
Accuracy:
tensor(0.9542, dtype=torch.float64)
Loss:
tensor(0.1890, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
14
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.1492, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
15
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.1552, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
16
Accuracy:
tensor(0.9583, dtype=torch.float64)
Loss:
tensor(0.1820, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
17
Accuracy:
tensor(0.9583, dtype=torch.float64)
Loss:
tensor(0.1711, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
18
Accuracy:
tensor(0.9542, dtype=torch.float64)
Loss:
tensor(0.2032, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
19
Accuracy:
tensor(0.9375, dtype=torch.float64)
Loss:
tensor(0.1943, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
20
########################
Validation Accuracy:
tensor(0.2817, dtype=torch.float64)
########################
Accuracy:
tensor(0.9458, dtype=torch.float64)
Loss:
tensor(0.1967, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
21
Accuracy:
tensor(0.9500, dtype=torch.float64)
Loss:
tensor(0.1859, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
22
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.1785, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
23
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.1623, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
24
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.1620, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
25
Accuracy:
tensor(0.9375, dtype=torch.float64)
Loss:
tensor(0.1761, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
26
Accuracy:
tensor(0.9625, dtype=torch.float64)
Loss:
tensor(0.1625, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
27
Accuracy:
tensor(0.9625, dtype=torch.float64)
Loss:
tensor(0.1747, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
28
Accuracy:
tensor(0.9583, dtype=torch.float64)
Loss:
tensor(0.2022, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
29
Accuracy:
tensor(0.9458, dtype=torch.float64)
Loss:
tensor(0.1785, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
30
Accuracy:
tensor(0.9625, dtype=torch.float64)
Loss:
tensor(0.1860, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
31
Accuracy:
tensor(0.9292, dtype=torch.float64)
Loss:
tensor(0.2016, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
32
Accuracy:
tensor(0.9375, dtype=torch.float64)
Loss:
tensor(0.2029, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
33
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.1687, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
34
Accuracy:
tensor(0.9458, dtype=torch.float64)
Loss:
tensor(0.1961, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
35
Accuracy:
tensor(0.9542, dtype=torch.float64)
Loss:
tensor(0.1720, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
36
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.1526, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
37
Accuracy:
tensor(0.9417, dtype=torch.float64)
Loss:
tensor(0.2189, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
38
Accuracy:
tensor(0.9500, dtype=torch.float64)
Loss:
tensor(0.2046, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
39
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.1557, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
40
########################
Validation Accuracy:
tensor(0.3050, dtype=torch.float64)
########################
Accuracy:
tensor(0.9542, dtype=torch.float64)
Loss:
tensor(0.1767, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
41
Accuracy:
tensor(0.9333, dtype=torch.float64)
Loss:
tensor(0.2290, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
42
Accuracy:
tensor(0.9250, dtype=torch.float64)
Loss:
tensor(0.2140, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
43
Accuracy:
tensor(0.9000, dtype=torch.float64)
Loss:
tensor(0.2624, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
44
Accuracy:
tensor(0.9458, dtype=torch.float64)
Loss:
tensor(0.2030, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
45
Accuracy:
tensor(0.8917, dtype=torch.float64)
Loss:
tensor(0.3061, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
46
Accuracy:
tensor(0.9583, dtype=torch.float64)
Loss:
tensor(0.1981, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
47
Accuracy:
tensor(0.9500, dtype=torch.float64)
Loss:
tensor(0.1798, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
48
Accuracy:
tensor(0.9542, dtype=torch.float64)
Loss:
tensor(0.1936, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
49
Accuracy:
tensor(0.9500, dtype=torch.float64)
Loss:
tensor(0.2082, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
50
Accuracy:
tensor(0.9333, dtype=torch.float64)
Loss:
tensor(0.2557, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
51
Accuracy:
tensor(0.9250, dtype=torch.float64)
Loss:
tensor(0.2292, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
52
Accuracy:
tensor(0.9208, dtype=torch.float64)
Loss:
tensor(0.2368, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
53
Accuracy:
tensor(0.9375, dtype=torch.float64)
Loss:
tensor(0.2182, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
54
Accuracy:
tensor(0.9292, dtype=torch.float64)
Loss:
tensor(0.2363, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
55
Accuracy:
tensor(0.9542, dtype=torch.float64)
Loss:
tensor(0.1826, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
56
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.1428, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
57
Accuracy:
tensor(0.9500, dtype=torch.float64)
Loss:
tensor(0.1964, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
58
Accuracy:
tensor(0.9292, dtype=torch.float64)
Loss:
tensor(0.2208, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
59
Accuracy:
tensor(0.9333, dtype=torch.float64)
Loss:
tensor(0.2206, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
60
########################
Validation Accuracy:
tensor(0.2933, dtype=torch.float64)
########################
Accuracy:
tensor(0.9583, dtype=torch.float64)
Loss:
tensor(0.1658, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
61
Accuracy:
tensor(0.9583, dtype=torch.float64)
Loss:
tensor(0.1694, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
62
Accuracy:
tensor(0.9583, dtype=torch.float64)
Loss:
tensor(0.1733, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
63
Accuracy:
tensor(0.9375, dtype=torch.float64)
Loss:
tensor(0.2419, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
64
Accuracy:
tensor(0.9583, dtype=torch.float64)
Loss:
tensor(0.1825, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
65
Accuracy:
tensor(0.9167, dtype=torch.float64)
Loss:
tensor(0.2429, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
66
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.1705, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
67
Accuracy:
tensor(0.9458, dtype=torch.float64)
Loss:
tensor(0.2006, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
68
Accuracy:
tensor(0.9458, dtype=torch.float64)
Loss:
tensor(0.1906, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
69
Accuracy:
tensor(0.9333, dtype=torch.float64)
Loss:
tensor(0.2234, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
70
Accuracy:
tensor(0.9417, dtype=torch.float64)
Loss:
tensor(0.2476, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
71
Accuracy:
tensor(0.9417, dtype=torch.float64)
Loss:
tensor(0.1941, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
72
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.1857, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
73
Accuracy:
tensor(0.9500, dtype=torch.float64)
Loss:
tensor(0.1923, grad_fn=<NllLossBackward>)
Epoch:
19
Batch:
74
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.1220, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
0
########################
Validation Accuracy:
tensor(0.3050, dtype=torch.float64)
########################
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.1346, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
1
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.1182, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
2
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.1329, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
3
Accuracy:
tensor(0.9583, dtype=torch.float64)
Loss:
tensor(0.1825, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
4
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.1394, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
5
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.1346, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
6
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.1406, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
7
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.1377, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
8
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.1294, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
9
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.1152, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
10
Accuracy:
tensor(0.9542, dtype=torch.float64)
Loss:
tensor(0.1571, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
11
Accuracy:
tensor(0.9625, dtype=torch.float64)
Loss:
tensor(0.1538, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
12
Accuracy:
tensor(0.9500, dtype=torch.float64)
Loss:
tensor(0.1948, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
13
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.1660, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
14
Accuracy:
tensor(0.9583, dtype=torch.float64)
Loss:
tensor(0.1577, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
15
Accuracy:
tensor(0.9458, dtype=torch.float64)
Loss:
tensor(0.1775, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
16
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.1130, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
17
Accuracy:
tensor(0.9625, dtype=torch.float64)
Loss:
tensor(0.1409, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
18
Accuracy:
tensor(0.9583, dtype=torch.float64)
Loss:
tensor(0.1554, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
19
Accuracy:
tensor(0.9542, dtype=torch.float64)
Loss:
tensor(0.1655, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
20
########################
Validation Accuracy:
tensor(0.3000, dtype=torch.float64)
########################
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.1506, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
21
Accuracy:
tensor(0.9542, dtype=torch.float64)
Loss:
tensor(0.1694, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
22
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.1306, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
23
Accuracy:
tensor(0.9333, dtype=torch.float64)
Loss:
tensor(0.1768, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
24
Accuracy:
tensor(0.9583, dtype=torch.float64)
Loss:
tensor(0.1928, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
25
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.1554, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
26
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.1423, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
27
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.1381, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
28
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.1431, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
29
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.1562, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
30
Accuracy:
tensor(0.9542, dtype=torch.float64)
Loss:
tensor(0.2022, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
31
Accuracy:
tensor(0.9542, dtype=torch.float64)
Loss:
tensor(0.1719, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
32
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.1374, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
33
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.1356, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
34
Accuracy:
tensor(0.9625, dtype=torch.float64)
Loss:
tensor(0.1546, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
35
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.1585, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
36
Accuracy:
tensor(0.9458, dtype=torch.float64)
Loss:
tensor(0.2102, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
37
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.1271, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
38
Accuracy:
tensor(0.9500, dtype=torch.float64)
Loss:
tensor(0.1545, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
39
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.1641, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
40
########################
Validation Accuracy:
tensor(0.3000, dtype=torch.float64)
########################
Accuracy:
tensor(0.9583, dtype=torch.float64)
Loss:
tensor(0.1562, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
41
Accuracy:
tensor(0.9292, dtype=torch.float64)
Loss:
tensor(0.2498, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
42
Accuracy:
tensor(0.9625, dtype=torch.float64)
Loss:
tensor(0.1595, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
43
Accuracy:
tensor(0.9375, dtype=torch.float64)
Loss:
tensor(0.1801, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
44
Accuracy:
tensor(0.9625, dtype=torch.float64)
Loss:
tensor(0.1498, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
45
Accuracy:
tensor(0.9417, dtype=torch.float64)
Loss:
tensor(0.1871, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
46
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.1643, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
47
Accuracy:
tensor(0.9250, dtype=torch.float64)
Loss:
tensor(0.2224, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
48
Accuracy:
tensor(0.9583, dtype=torch.float64)
Loss:
tensor(0.1929, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
49
Accuracy:
tensor(0.9458, dtype=torch.float64)
Loss:
tensor(0.1962, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
50
Accuracy:
tensor(0.9375, dtype=torch.float64)
Loss:
tensor(0.1679, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
51
Accuracy:
tensor(0.9500, dtype=torch.float64)
Loss:
tensor(0.1761, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
52
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.1105, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
53
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.1681, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
54
Accuracy:
tensor(0.9542, dtype=torch.float64)
Loss:
tensor(0.1976, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
55
Accuracy:
tensor(0.9333, dtype=torch.float64)
Loss:
tensor(0.1854, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
56
Accuracy:
tensor(0.9583, dtype=torch.float64)
Loss:
tensor(0.1523, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
57
Accuracy:
tensor(0.9458, dtype=torch.float64)
Loss:
tensor(0.1690, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
58
Accuracy:
tensor(0.9542, dtype=torch.float64)
Loss:
tensor(0.1752, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
59
Accuracy:
tensor(0.9583, dtype=torch.float64)
Loss:
tensor(0.1417, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
60
########################
Validation Accuracy:
tensor(0.3133, dtype=torch.float64)
########################
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.1683, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
61
Accuracy:
tensor(0.9458, dtype=torch.float64)
Loss:
tensor(0.1855, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
62
Accuracy:
tensor(0.9500, dtype=torch.float64)
Loss:
tensor(0.1664, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
63
Accuracy:
tensor(0.9625, dtype=torch.float64)
Loss:
tensor(0.1852, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
64
Accuracy:
tensor(0.9542, dtype=torch.float64)
Loss:
tensor(0.1655, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
65
Accuracy:
tensor(0.9333, dtype=torch.float64)
Loss:
tensor(0.2064, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
66
Accuracy:
tensor(0.9625, dtype=torch.float64)
Loss:
tensor(0.1465, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
67
Accuracy:
tensor(0.9583, dtype=torch.float64)
Loss:
tensor(0.1746, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
68
Accuracy:
tensor(0.9333, dtype=torch.float64)
Loss:
tensor(0.1805, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
69
Accuracy:
tensor(0.9542, dtype=torch.float64)
Loss:
tensor(0.1640, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
70
Accuracy:
tensor(0.9542, dtype=torch.float64)
Loss:
tensor(0.2016, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
71
Accuracy:
tensor(0.9333, dtype=torch.float64)
Loss:
tensor(0.2161, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
72
Accuracy:
tensor(0.9500, dtype=torch.float64)
Loss:
tensor(0.1775, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
73
Accuracy:
tensor(0.9542, dtype=torch.float64)
Loss:
tensor(0.1832, grad_fn=<NllLossBackward>)
Epoch:
20
Batch:
74
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.1185, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
0
########################
Validation Accuracy:
tensor(0.3067, dtype=torch.float64)
########################
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.1105, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
1
Accuracy:
tensor(0.9542, dtype=torch.float64)
Loss:
tensor(0.1468, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
2
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.1237, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
3
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.1233, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
4
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.1348, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
5
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.1091, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
6
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.1079, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
7
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.1340, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
8
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.1246, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
9
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.1400, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
10
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.1007, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
11
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.1016, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
12
Accuracy:
tensor(0.9625, dtype=torch.float64)
Loss:
tensor(0.1357, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
13
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.1322, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
14
Accuracy:
tensor(0.9583, dtype=torch.float64)
Loss:
tensor(0.1307, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
15
Accuracy:
tensor(0.9583, dtype=torch.float64)
Loss:
tensor(0.1439, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
16
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.1480, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
17
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.1127, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
18
Accuracy:
tensor(0.9625, dtype=torch.float64)
Loss:
tensor(0.1289, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
19
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.1164, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
20
########################
Validation Accuracy:
tensor(0.3067, dtype=torch.float64)
########################
Accuracy:
tensor(0.9625, dtype=torch.float64)
Loss:
tensor(0.1250, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
21
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.1261, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
22
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.1146, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
23
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.1341, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
24
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.1185, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
25
Accuracy:
tensor(0.9625, dtype=torch.float64)
Loss:
tensor(0.1565, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
26
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.1047, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
27
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.1384, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
28
Accuracy:
tensor(0.9458, dtype=torch.float64)
Loss:
tensor(0.1438, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
29
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.1242, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
30
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.1363, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
31
Accuracy:
tensor(0.9583, dtype=torch.float64)
Loss:
tensor(0.1343, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
32
Accuracy:
tensor(0.9625, dtype=torch.float64)
Loss:
tensor(0.1637, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
33
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.1290, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
34
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.1356, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
35
Accuracy:
tensor(0.9542, dtype=torch.float64)
Loss:
tensor(0.1417, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
36
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.1198, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
37
Accuracy:
tensor(0.9625, dtype=torch.float64)
Loss:
tensor(0.1352, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
38
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.1290, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
39
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.1314, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
40
########################
Validation Accuracy:
tensor(0.3000, dtype=torch.float64)
########################
Accuracy:
tensor(0.9625, dtype=torch.float64)
Loss:
tensor(0.1451, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
41
Accuracy:
tensor(0.9625, dtype=torch.float64)
Loss:
tensor(0.1480, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
42
Accuracy:
tensor(0.9583, dtype=torch.float64)
Loss:
tensor(0.1761, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
43
Accuracy:
tensor(0.9417, dtype=torch.float64)
Loss:
tensor(0.1723, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
44
Accuracy:
tensor(0.9458, dtype=torch.float64)
Loss:
tensor(0.1982, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
45
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.1367, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
46
Accuracy:
tensor(0.9583, dtype=torch.float64)
Loss:
tensor(0.1387, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
47
Accuracy:
tensor(0.9417, dtype=torch.float64)
Loss:
tensor(0.1596, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
48
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.1206, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
49
Accuracy:
tensor(0.9583, dtype=torch.float64)
Loss:
tensor(0.1306, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
50
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.1374, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
51
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.1381, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
52
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.1134, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
53
Accuracy:
tensor(0.9542, dtype=torch.float64)
Loss:
tensor(0.1762, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
54
Accuracy:
tensor(0.9417, dtype=torch.float64)
Loss:
tensor(0.2194, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
55
Accuracy:
tensor(0.9458, dtype=torch.float64)
Loss:
tensor(0.1667, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
56
Accuracy:
tensor(0.9417, dtype=torch.float64)
Loss:
tensor(0.1884, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
57
Accuracy:
tensor(0.9625, dtype=torch.float64)
Loss:
tensor(0.1589, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
58
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.1242, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
59
Accuracy:
tensor(0.9583, dtype=torch.float64)
Loss:
tensor(0.1546, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
60
########################
Validation Accuracy:
tensor(0.3017, dtype=torch.float64)
########################
Accuracy:
tensor(0.9625, dtype=torch.float64)
Loss:
tensor(0.1538, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
61
Accuracy:
tensor(0.9625, dtype=torch.float64)
Loss:
tensor(0.1458, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
62
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.1427, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
63
Accuracy:
tensor(0.9583, dtype=torch.float64)
Loss:
tensor(0.1434, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
64
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.1416, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
65
Accuracy:
tensor(0.9583, dtype=torch.float64)
Loss:
tensor(0.1656, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
66
Accuracy:
tensor(0.9542, dtype=torch.float64)
Loss:
tensor(0.1722, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
67
Accuracy:
tensor(0.9625, dtype=torch.float64)
Loss:
tensor(0.1437, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
68
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.1099, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
69
Accuracy:
tensor(0.9583, dtype=torch.float64)
Loss:
tensor(0.1713, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
70
Accuracy:
tensor(0.9458, dtype=torch.float64)
Loss:
tensor(0.1716, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
71
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.1396, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
72
Accuracy:
tensor(0.9542, dtype=torch.float64)
Loss:
tensor(0.2057, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
73
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.1216, grad_fn=<NllLossBackward>)
Epoch:
21
Batch:
74
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0990, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
0
########################
Validation Accuracy:
tensor(0.2883, dtype=torch.float64)
########################
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.1197, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
1
Accuracy:
tensor(0.9542, dtype=torch.float64)
Loss:
tensor(0.1714, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
2
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0967, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
3
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0835, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
4
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.1382, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
5
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.1329, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
6
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.1234, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
7
Accuracy:
tensor(0.9625, dtype=torch.float64)
Loss:
tensor(0.1565, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
8
Accuracy:
tensor(0.9583, dtype=torch.float64)
Loss:
tensor(0.1519, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
9
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.1044, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
10
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.1054, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
11
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.1222, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
12
Accuracy:
tensor(0.9583, dtype=torch.float64)
Loss:
tensor(0.1495, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
13
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.1230, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
14
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0988, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
15
Accuracy:
tensor(0.9458, dtype=torch.float64)
Loss:
tensor(0.1587, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
16
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0949, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
17
Accuracy:
tensor(0.9625, dtype=torch.float64)
Loss:
tensor(0.1331, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
18
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.1280, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
19
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.1124, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
20
########################
Validation Accuracy:
tensor(0.2917, dtype=torch.float64)
########################
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.1143, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
21
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.1175, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
22
Accuracy:
tensor(0.9500, dtype=torch.float64)
Loss:
tensor(0.1787, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
23
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0897, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
24
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.1033, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
25
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.1212, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
26
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.1170, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
27
Accuracy:
tensor(0.9500, dtype=torch.float64)
Loss:
tensor(0.1508, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
28
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.1091, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
29
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0858, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
30
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.1048, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
31
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0936, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
32
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.1154, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
33
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.1421, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
34
Accuracy:
tensor(0.9625, dtype=torch.float64)
Loss:
tensor(0.1389, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
35
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.1159, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
36
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.1147, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
37
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.1290, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
38
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.1042, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
39
Accuracy:
tensor(0.9542, dtype=torch.float64)
Loss:
tensor(0.1475, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
40
########################
Validation Accuracy:
tensor(0.2917, dtype=torch.float64)
########################
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0873, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
41
Accuracy:
tensor(0.9583, dtype=torch.float64)
Loss:
tensor(0.1671, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
42
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.1331, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
43
Accuracy:
tensor(0.9542, dtype=torch.float64)
Loss:
tensor(0.1315, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
44
Accuracy:
tensor(0.9542, dtype=torch.float64)
Loss:
tensor(0.1391, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
45
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.1245, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
46
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.1468, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
47
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0745, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
48
Accuracy:
tensor(0.9542, dtype=torch.float64)
Loss:
tensor(0.1426, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
49
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.1199, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
50
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.1351, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
51
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.1017, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
52
Accuracy:
tensor(0.9625, dtype=torch.float64)
Loss:
tensor(0.1287, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
53
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.1144, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
54
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.1350, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
55
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.1289, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
56
Accuracy:
tensor(0.9542, dtype=torch.float64)
Loss:
tensor(0.1529, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
57
Accuracy:
tensor(0.9583, dtype=torch.float64)
Loss:
tensor(0.1601, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
58
Accuracy:
tensor(0.9500, dtype=torch.float64)
Loss:
tensor(0.1426, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
59
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0849, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
60
########################
Validation Accuracy:
tensor(0.2917, dtype=torch.float64)
########################
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0858, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
61
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.1260, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
62
Accuracy:
tensor(0.9542, dtype=torch.float64)
Loss:
tensor(0.1735, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
63
Accuracy:
tensor(0.9625, dtype=torch.float64)
Loss:
tensor(0.1163, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
64
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.1401, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
65
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.1426, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
66
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.1306, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
67
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.1330, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
68
Accuracy:
tensor(0.9542, dtype=torch.float64)
Loss:
tensor(0.1545, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
69
Accuracy:
tensor(0.9583, dtype=torch.float64)
Loss:
tensor(0.1434, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
70
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.1294, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
71
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.1436, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
72
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.1441, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
73
Accuracy:
tensor(0.9500, dtype=torch.float64)
Loss:
tensor(0.1754, grad_fn=<NllLossBackward>)
Epoch:
22
Batch:
74
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0943, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
0
########################
Validation Accuracy:
tensor(0.2967, dtype=torch.float64)
########################
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0836, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
1
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0719, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
2
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0911, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
3
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.1026, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
4
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0981, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
5
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0920, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
6
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.1097, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
7
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0833, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
8
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.0985, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
9
Accuracy:
tensor(0.9542, dtype=torch.float64)
Loss:
tensor(0.1137, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
10
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0947, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
11
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0725, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
12
Accuracy:
tensor(0.9625, dtype=torch.float64)
Loss:
tensor(0.1214, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
13
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0986, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
14
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0898, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
15
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0894, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
16
Accuracy:
tensor(0.9625, dtype=torch.float64)
Loss:
tensor(0.1283, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
17
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0910, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
18
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0852, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
19
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0836, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
20
########################
Validation Accuracy:
tensor(0.2983, dtype=torch.float64)
########################
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0992, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
21
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.1036, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
22
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0998, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
23
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0966, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
24
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.1298, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
25
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0858, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
26
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0846, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
27
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0820, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
28
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.1030, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
29
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0760, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
30
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.1015, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
31
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.1071, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
32
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.1499, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
33
Accuracy:
tensor(0.9583, dtype=torch.float64)
Loss:
tensor(0.1307, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
34
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.1138, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
35
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0763, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
36
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.1026, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
37
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.1125, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
38
Accuracy:
tensor(0.9542, dtype=torch.float64)
Loss:
tensor(0.1476, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
39
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.1189, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
40
########################
Validation Accuracy:
tensor(0.2983, dtype=torch.float64)
########################
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.1171, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
41
Accuracy:
tensor(0.9625, dtype=torch.float64)
Loss:
tensor(0.1308, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
42
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.1090, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
43
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.1091, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
44
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.1261, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
45
Accuracy:
tensor(0.9583, dtype=torch.float64)
Loss:
tensor(0.1400, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
46
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0795, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
47
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.1059, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
48
Accuracy:
tensor(0.9625, dtype=torch.float64)
Loss:
tensor(0.1301, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
49
Accuracy:
tensor(0.9542, dtype=torch.float64)
Loss:
tensor(0.1758, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
50
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.1284, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
51
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.1205, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
52
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.1320, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
53
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0882, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
54
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0975, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
55
Accuracy:
tensor(0.9500, dtype=torch.float64)
Loss:
tensor(0.1452, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
56
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.1198, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
57
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.1258, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
58
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.1082, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
59
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.1202, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
60
########################
Validation Accuracy:
tensor(0.3050, dtype=torch.float64)
########################
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.1331, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
61
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.1107, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
62
Accuracy:
tensor(0.9625, dtype=torch.float64)
Loss:
tensor(0.1526, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
63
Accuracy:
tensor(0.9625, dtype=torch.float64)
Loss:
tensor(0.1294, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
64
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.0964, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
65
Accuracy:
tensor(0.9458, dtype=torch.float64)
Loss:
tensor(0.1479, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
66
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.1388, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
67
Accuracy:
tensor(0.9500, dtype=torch.float64)
Loss:
tensor(0.1583, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
68
Accuracy:
tensor(0.9583, dtype=torch.float64)
Loss:
tensor(0.1244, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
69
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.1540, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
70
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.1021, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
71
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0936, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
72
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.1176, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
73
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.1073, grad_fn=<NllLossBackward>)
Epoch:
23
Batch:
74
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0733, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
0
########################
Validation Accuracy:
tensor(0.2933, dtype=torch.float64)
########################
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.1008, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
1
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0868, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
2
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0574, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
3
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0895, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
4
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0766, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
5
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.1018, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
6
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0718, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
7
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0827, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
8
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0833, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
9
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0825, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
10
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0764, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
11
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0766, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
12
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0749, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
13
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0850, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
14
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0859, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
15
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0602, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
16
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0718, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
17
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0903, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
18
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.1170, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
19
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0745, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
20
########################
Validation Accuracy:
tensor(0.2867, dtype=torch.float64)
########################
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0916, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
21
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.1076, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
22
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0742, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
23
Accuracy:
tensor(0.9583, dtype=torch.float64)
Loss:
tensor(0.1359, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
24
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.1213, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
25
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0756, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
26
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0882, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
27
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.0942, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
28
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0783, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
29
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0817, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
30
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0956, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
31
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.1022, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
32
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0845, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
33
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0998, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
34
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.1075, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
35
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0674, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
36
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0945, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
37
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.1043, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
38
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0887, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
39
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0781, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
40
########################
Validation Accuracy:
tensor(0.2967, dtype=torch.float64)
########################
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0901, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
41
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0948, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
42
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.1314, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
43
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0898, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
44
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.1028, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
45
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.1202, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
46
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0605, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
47
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.1083, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
48
Accuracy:
tensor(0.9625, dtype=torch.float64)
Loss:
tensor(0.1148, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
49
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.1025, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
50
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.1068, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
51
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.1031, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
52
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.1185, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
53
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0609, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
54
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0984, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
55
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.1009, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
56
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0943, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
57
Accuracy:
tensor(0.9500, dtype=torch.float64)
Loss:
tensor(0.1479, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
58
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.1047, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
59
Accuracy:
tensor(0.9500, dtype=torch.float64)
Loss:
tensor(0.1232, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
60
########################
Validation Accuracy:
tensor(0.3083, dtype=torch.float64)
########################
Accuracy:
tensor(0.9500, dtype=torch.float64)
Loss:
tensor(0.1428, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
61
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.1535, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
62
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.1067, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
63
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.1168, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
64
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.1023, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
65
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.1009, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
66
Accuracy:
tensor(0.9500, dtype=torch.float64)
Loss:
tensor(0.1667, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
67
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.1020, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
68
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0736, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
69
Accuracy:
tensor(0.9583, dtype=torch.float64)
Loss:
tensor(0.1367, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
70
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.1154, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
71
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0860, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
72
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.1068, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
73
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0769, grad_fn=<NllLossBackward>)
Epoch:
24
Batch:
74
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0540, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
0
########################
Validation Accuracy:
tensor(0.2933, dtype=torch.float64)
########################
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0634, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
1
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0528, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
2
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0906, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
3
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0654, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
4
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0653, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
5
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0668, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
6
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0914, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
7
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.1208, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
8
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.1036, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
9
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0832, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
10
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0643, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
11
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.1097, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
12
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0530, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
13
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0638, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
14
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0822, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
15
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0843, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
16
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0670, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
17
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.1036, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
18
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0722, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
19
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.1038, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
20
########################
Validation Accuracy:
tensor(0.3117, dtype=torch.float64)
########################
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0671, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
21
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0779, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
22
Accuracy:
tensor(0.9542, dtype=torch.float64)
Loss:
tensor(0.1336, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
23
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0629, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
24
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.1152, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
25
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.1239, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
26
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0852, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
27
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0568, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
28
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0850, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
29
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0973, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
30
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0939, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
31
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0657, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
32
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0625, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
33
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0850, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
34
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0720, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
35
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0657, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
36
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0639, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
37
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0643, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
38
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0645, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
39
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0815, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
40
########################
Validation Accuracy:
tensor(0.3083, dtype=torch.float64)
########################
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0711, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
41
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0821, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
42
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0796, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
43
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.1234, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
44
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0726, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
45
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0755, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
46
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0881, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
47
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0629, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
48
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0787, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
49
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.1006, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
50
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0998, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
51
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.1076, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
52
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0652, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
53
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0803, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
54
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0923, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
55
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0703, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
56
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0644, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
57
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0984, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
58
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0688, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
59
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0803, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
60
########################
Validation Accuracy:
tensor(0.3050, dtype=torch.float64)
########################
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0683, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
61
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.0983, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
62
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0713, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
63
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0844, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
64
Accuracy:
tensor(0.9625, dtype=torch.float64)
Loss:
tensor(0.1004, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
65
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0880, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
66
Accuracy:
tensor(0.9583, dtype=torch.float64)
Loss:
tensor(0.1290, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
67
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0989, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
68
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0758, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
69
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0859, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
70
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0932, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
71
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0838, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
72
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0708, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
73
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0781, grad_fn=<NllLossBackward>)
Epoch:
25
Batch:
74
Accuracy:
tensor(0.9625, dtype=torch.float64)
Loss:
tensor(0.0907, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
0
########################
Validation Accuracy:
tensor(0.3117, dtype=torch.float64)
########################
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0825, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
1
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0799, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
2
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0453, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
3
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0571, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
4
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0704, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
5
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.0808, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
6
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0692, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
7
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0650, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
8
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0797, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
9
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0560, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
10
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0434, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
11
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0676, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
12
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0759, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
13
Accuracy:
tensor(0.9625, dtype=torch.float64)
Loss:
tensor(0.1194, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
14
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0654, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
15
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0842, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
16
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0535, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
17
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0562, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
18
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0853, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
19
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0775, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
20
########################
Validation Accuracy:
tensor(0.3100, dtype=torch.float64)
########################
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0707, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
21
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0629, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
22
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0546, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
23
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0766, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
24
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0521, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
25
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0774, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
26
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0993, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
27
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0809, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
28
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0595, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
29
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0866, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
30
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0735, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
31
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0960, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
32
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0897, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
33
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.1083, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
34
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0742, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
35
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0696, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
36
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0610, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
37
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0792, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
38
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0643, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
39
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0381, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
40
########################
Validation Accuracy:
tensor(0.2900, dtype=torch.float64)
########################
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0834, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
41
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0799, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
42
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0591, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
43
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0793, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
44
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0748, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
45
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0604, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
46
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0861, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
47
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0775, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
48
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.1025, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
49
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0773, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
50
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0967, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
51
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.1016, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
52
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0715, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
53
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0713, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
54
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.0992, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
55
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0923, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
56
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0586, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
57
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0753, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
58
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0883, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
59
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0650, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
60
########################
Validation Accuracy:
tensor(0.3050, dtype=torch.float64)
########################
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0640, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
61
Accuracy:
tensor(0.9583, dtype=torch.float64)
Loss:
tensor(0.1335, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
62
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0636, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
63
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0650, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
64
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0995, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
65
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.1011, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
66
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0713, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
67
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0904, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
68
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.0948, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
69
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.1162, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
70
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0831, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
71
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0924, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
72
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0636, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
73
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0868, grad_fn=<NllLossBackward>)
Epoch:
26
Batch:
74
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0568, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
0
########################
Validation Accuracy:
tensor(0.2950, dtype=torch.float64)
########################
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0536, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
1
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0701, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
2
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0533, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
3
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0575, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
4
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0583, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
5
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0621, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
6
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0643, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
7
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0572, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
8
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0834, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
9
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0382, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
10
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0676, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
11
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0674, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
12
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0519, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
13
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0646, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
14
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0808, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
15
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0607, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
16
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0661, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
17
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0602, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
18
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0655, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
19
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0555, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
20
########################
Validation Accuracy:
tensor(0.2950, dtype=torch.float64)
########################
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0674, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
21
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0698, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
22
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0478, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
23
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0628, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
24
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.1035, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
25
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0705, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
26
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0676, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
27
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0822, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
28
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0513, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
29
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0607, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
30
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0635, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
31
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0997, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
32
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0781, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
33
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0459, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
34
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0795, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
35
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0876, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
36
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0751, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
37
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0648, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
38
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0639, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
39
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0618, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
40
########################
Validation Accuracy:
tensor(0.2950, dtype=torch.float64)
########################
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0853, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
41
Accuracy:
tensor(0.9542, dtype=torch.float64)
Loss:
tensor(0.1518, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
42
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0598, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
43
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0571, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
44
Accuracy:
tensor(0.9500, dtype=torch.float64)
Loss:
tensor(0.1438, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
45
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0829, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
46
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0714, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
47
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.1096, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
48
Accuracy:
tensor(0.9625, dtype=torch.float64)
Loss:
tensor(0.1203, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
49
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.0957, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
50
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0758, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
51
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0806, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
52
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0722, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
53
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.1166, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
54
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0757, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
55
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.1030, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
56
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0757, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
57
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0782, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
58
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0829, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
59
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0678, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
60
########################
Validation Accuracy:
tensor(0.3167, dtype=torch.float64)
########################
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0625, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
61
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0809, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
62
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0824, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
63
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0939, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
64
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0926, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
65
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0878, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
66
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0668, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
67
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0749, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
68
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0579, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
69
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0715, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
70
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0848, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
71
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0655, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
72
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0979, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
73
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.1266, grad_fn=<NllLossBackward>)
Epoch:
27
Batch:
74
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0423, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
0
########################
Validation Accuracy:
tensor(0.2933, dtype=torch.float64)
########################
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0517, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
1
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0717, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
2
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0700, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
3
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0688, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
4
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0473, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
5
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0534, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
6
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0572, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
7
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0458, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
8
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0560, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
9
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0531, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
10
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0406, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
11
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0572, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
12
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0626, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
13
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0585, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
14
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0540, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
15
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0948, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
16
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0853, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
17
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0444, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
18
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0391, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
19
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0502, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
20
########################
Validation Accuracy:
tensor(0.2883, dtype=torch.float64)
########################
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0587, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
21
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0782, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
22
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0571, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
23
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0676, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
24
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0761, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
25
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0482, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
26
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0561, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
27
Accuracy:
tensor(0.9625, dtype=torch.float64)
Loss:
tensor(0.1053, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
28
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0672, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
29
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0968, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
30
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0669, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
31
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0474, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
32
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0739, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
33
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0685, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
34
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0819, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
35
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0524, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
36
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0666, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
37
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0598, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
38
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0753, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
39
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0688, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
40
########################
Validation Accuracy:
tensor(0.3050, dtype=torch.float64)
########################
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0657, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
41
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0749, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
42
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0769, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
43
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0667, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
44
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0655, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
45
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0666, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
46
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0639, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
47
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0757, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
48
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0563, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
49
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0735, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
50
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0685, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
51
Accuracy:
tensor(0.9625, dtype=torch.float64)
Loss:
tensor(0.1157, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
52
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0576, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
53
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0843, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
54
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0518, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
55
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0660, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
56
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0633, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
57
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0681, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
58
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0580, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
59
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0472, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
60
########################
Validation Accuracy:
tensor(0.3150, dtype=torch.float64)
########################
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0505, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
61
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0493, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
62
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0727, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
63
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0846, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
64
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0721, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
65
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0541, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
66
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.0887, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
67
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0997, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
68
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0897, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
69
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.1148, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
70
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0849, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
71
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0901, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
72
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0806, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
73
Accuracy:
tensor(0.9625, dtype=torch.float64)
Loss:
tensor(0.1120, grad_fn=<NllLossBackward>)
Epoch:
28
Batch:
74
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0445, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
0
########################
Validation Accuracy:
tensor(0.2950, dtype=torch.float64)
########################
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0518, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
1
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0510, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
2
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0874, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
3
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0582, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
4
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0406, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
5
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0683, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
6
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0792, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
7
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0466, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
8
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0527, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
9
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0454, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
10
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0504, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
11
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0792, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
12
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0554, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
13
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0471, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
14
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0516, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
15
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0418, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
16
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0399, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
17
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0547, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
18
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0499, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
19
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0578, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
20
########################
Validation Accuracy:
tensor(0.3050, dtype=torch.float64)
########################
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0481, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
21
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0635, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
22
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0489, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
23
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0730, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
24
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0642, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
25
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0519, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
26
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0349, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
27
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0433, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
28
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0701, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
29
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0561, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
30
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0403, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
31
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0558, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
32
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0652, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
33
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0670, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
34
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0674, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
35
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0587, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
36
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0583, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
37
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0639, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
38
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0437, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
39
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0853, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
40
########################
Validation Accuracy:
tensor(0.3050, dtype=torch.float64)
########################
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0650, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
41
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0690, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
42
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0564, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
43
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0436, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
44
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0614, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
45
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0596, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
46
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0626, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
47
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0483, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
48
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0777, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
49
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0537, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
50
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0479, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
51
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0716, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
52
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0552, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
53
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0643, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
54
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0395, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
55
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0878, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
56
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0876, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
57
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0750, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
58
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0656, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
59
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0625, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
60
########################
Validation Accuracy:
tensor(0.3317, dtype=torch.float64)
########################
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0706, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
61
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0544, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
62
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.1013, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
63
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0778, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
64
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0496, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
65
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0652, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
66
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0574, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
67
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0832, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
68
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0763, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
69
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0608, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
70
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0885, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
71
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0829, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
72
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0727, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
73
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0649, grad_fn=<NllLossBackward>)
Epoch:
29
Batch:
74
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0332, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
0
########################
Validation Accuracy:
tensor(0.3033, dtype=torch.float64)
########################
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0387, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
1
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0516, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
2
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0274, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
3
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0498, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
4
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0428, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
5
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0764, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
6
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0567, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
7
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0438, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
8
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0503, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
9
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0337, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
10
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0507, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
11
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0403, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
12
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0319, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
13
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0441, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
14
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0467, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
15
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0436, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
16
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0548, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
17
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0406, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
18
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0465, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
19
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0568, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
20
########################
Validation Accuracy:
tensor(0.3017, dtype=torch.float64)
########################
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0414, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
21
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0401, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
22
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0542, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
23
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0560, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
24
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0452, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
25
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0503, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
26
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0675, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
27
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0556, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
28
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0514, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
29
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0599, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
30
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0733, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
31
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0561, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
32
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0394, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
33
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0447, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
34
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0457, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
35
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0477, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
36
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0492, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
37
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0379, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
38
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0556, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
39
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0377, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
40
########################
Validation Accuracy:
tensor(0.3017, dtype=torch.float64)
########################
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0585, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
41
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0370, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
42
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0407, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
43
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0460, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
44
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0600, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
45
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0929, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
46
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0448, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
47
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0806, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
48
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0446, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
49
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0607, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
50
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0987, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
51
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0362, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
52
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0601, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
53
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0682, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
54
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0563, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
55
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0772, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
56
Accuracy:
tensor(0.9583, dtype=torch.float64)
Loss:
tensor(0.0970, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
57
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0687, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
58
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0704, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
59
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.1152, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
60
########################
Validation Accuracy:
tensor(0.2917, dtype=torch.float64)
########################
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0479, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
61
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.0745, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
62
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0470, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
63
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0917, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
64
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0739, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
65
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0554, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
66
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0941, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
67
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0624, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
68
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0628, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
69
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0727, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
70
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0777, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
71
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0551, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
72
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0483, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
73
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0599, grad_fn=<NllLossBackward>)
Epoch:
30
Batch:
74
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0463, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
0
########################
Validation Accuracy:
tensor(0.3000, dtype=torch.float64)
########################
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0674, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
1
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0379, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
2
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0453, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
3
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0788, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
4
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0499, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
5
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0640, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
6
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0730, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
7
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0475, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
8
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0407, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
9
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0314, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
10
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0409, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
11
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0447, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
12
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0600, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
13
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0588, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
14
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0688, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
15
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0462, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
16
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0460, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
17
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0395, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
18
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0427, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
19
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0626, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
20
########################
Validation Accuracy:
tensor(0.3100, dtype=torch.float64)
########################
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0546, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
21
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0423, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
22
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0523, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
23
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0317, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
24
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0815, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
25
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0602, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
26
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0631, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
27
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0556, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
28
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0522, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
29
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0573, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
30
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0331, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
31
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0495, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
32
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0457, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
33
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0393, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
34
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0615, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
35
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0551, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
36
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0497, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
37
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0693, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
38
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0478, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
39
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0442, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
40
########################
Validation Accuracy:
tensor(0.3117, dtype=torch.float64)
########################
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0595, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
41
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0533, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
42
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0429, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
43
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0578, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
44
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0412, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
45
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0536, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
46
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0398, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
47
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0471, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
48
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0466, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
49
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0493, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
50
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0437, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
51
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0493, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
52
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0551, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
53
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0286, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
54
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0859, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
55
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0455, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
56
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0604, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
57
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0589, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
58
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0661, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
59
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0490, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
60
########################
Validation Accuracy:
tensor(0.3083, dtype=torch.float64)
########################
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0554, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
61
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0619, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
62
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0674, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
63
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0581, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
64
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0437, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
65
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0473, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
66
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0459, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
67
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0513, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
68
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0652, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
69
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0606, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
70
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0549, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
71
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0587, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
72
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0526, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
73
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0418, grad_fn=<NllLossBackward>)
Epoch:
31
Batch:
74
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0432, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
0
########################
Validation Accuracy:
tensor(0.2983, dtype=torch.float64)
########################
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0346, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
1
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0425, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
2
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0298, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
3
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0464, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
4
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0387, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
5
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0483, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
6
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0692, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
7
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0600, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
8
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0331, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
9
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0558, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
10
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0291, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
11
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0316, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
12
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0343, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
13
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0205, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
14
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0337, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
15
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0424, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
16
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0405, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
17
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0254, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
18
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0445, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
19
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0384, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
20
########################
Validation Accuracy:
tensor(0.2983, dtype=torch.float64)
########################
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0375, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
21
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0534, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
22
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0326, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
23
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0530, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
24
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0260, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
25
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0434, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
26
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0299, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
27
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0507, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
28
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0428, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
29
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0465, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
30
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0565, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
31
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0416, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
32
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0349, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
33
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0323, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
34
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0328, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
35
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0313, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
36
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0289, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
37
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0217, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
38
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0660, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
39
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0289, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
40
########################
Validation Accuracy:
tensor(0.3200, dtype=torch.float64)
########################
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0491, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
41
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0271, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
42
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0540, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
43
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0629, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
44
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0513, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
45
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0447, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
46
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0400, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
47
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0557, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
48
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0410, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
49
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0421, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
50
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0309, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
51
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0598, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
52
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0713, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
53
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0600, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
54
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0664, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
55
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0443, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
56
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0469, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
57
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0492, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
58
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0502, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
59
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0492, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
60
########################
Validation Accuracy:
tensor(0.2900, dtype=torch.float64)
########################
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0550, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
61
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0344, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
62
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0546, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
63
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0476, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
64
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0452, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
65
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0425, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
66
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0350, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
67
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0325, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
68
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0472, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
69
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0494, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
70
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0694, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
71
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0330, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
72
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0499, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
73
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0371, grad_fn=<NllLossBackward>)
Epoch:
32
Batch:
74
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0293, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
0
########################
Validation Accuracy:
tensor(0.2983, dtype=torch.float64)
########################
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0306, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
1
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0429, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
2
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0379, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
3
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0300, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
4
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0554, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
5
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0376, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
6
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0231, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
7
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0508, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
8
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0437, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
9
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0334, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
10
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0321, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
11
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0385, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
12
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0217, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
13
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0292, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
14
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0438, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
15
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0214, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
16
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0378, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
17
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0285, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
18
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0498, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
19
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0295, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
20
########################
Validation Accuracy:
tensor(0.3067, dtype=torch.float64)
########################
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0471, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
21
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0410, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
22
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0270, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
23
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0536, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
24
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0290, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
25
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0315, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
26
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0463, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
27
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0456, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
28
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0414, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
29
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0416, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
30
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0301, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
31
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0500, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
32
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0564, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
33
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0445, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
34
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0535, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
35
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0388, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
36
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0561, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
37
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0369, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
38
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0535, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
39
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0650, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
40
########################
Validation Accuracy:
tensor(0.2967, dtype=torch.float64)
########################
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0442, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
41
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0397, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
42
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0528, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
43
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0517, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
44
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0386, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
45
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0220, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
46
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0654, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
47
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0430, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
48
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0276, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
49
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0424, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
50
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0508, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
51
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0402, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
52
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0402, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
53
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0600, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
54
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0362, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
55
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0297, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
56
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0335, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
57
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0514, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
58
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0465, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
59
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0485, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
60
########################
Validation Accuracy:
tensor(0.2950, dtype=torch.float64)
########################
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0518, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
61
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0365, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
62
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0340, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
63
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0554, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
64
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0578, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
65
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0393, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
66
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0325, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
67
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0462, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
68
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0623, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
69
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0391, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
70
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0748, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
71
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0736, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
72
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0580, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
73
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0515, grad_fn=<NllLossBackward>)
Epoch:
33
Batch:
74
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0388, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
0
########################
Validation Accuracy:
tensor(0.3000, dtype=torch.float64)
########################
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0315, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
1
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0317, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
2
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0529, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
3
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0288, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
4
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0376, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
5
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0724, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
6
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0480, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
7
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0278, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
8
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0243, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
9
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0587, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
10
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0342, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
11
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0390, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
12
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0224, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
13
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0345, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
14
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0319, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
15
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0499, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
16
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0586, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
17
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0614, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
18
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0308, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
19
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0504, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
20
########################
Validation Accuracy:
tensor(0.2933, dtype=torch.float64)
########################
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0351, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
21
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0534, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
22
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0421, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
23
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0495, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
24
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0498, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
25
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0196, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
26
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0317, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
27
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0585, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
28
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0505, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
29
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0520, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
30
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0335, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
31
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0573, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
32
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0510, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
33
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0322, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
34
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0341, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
35
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0418, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
36
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0470, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
37
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0319, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
38
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0452, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
39
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0414, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
40
########################
Validation Accuracy:
tensor(0.3050, dtype=torch.float64)
########################
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0398, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
41
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0554, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
42
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0612, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
43
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0566, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
44
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0409, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
45
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0595, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
46
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0595, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
47
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0203, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
48
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0443, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
49
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0369, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
50
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0297, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
51
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0426, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
52
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0554, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
53
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0479, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
54
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0452, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
55
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0424, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
56
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0541, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
57
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0472, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
58
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0439, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
59
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0191, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
60
########################
Validation Accuracy:
tensor(0.2917, dtype=torch.float64)
########################
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0792, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
61
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0677, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
62
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0434, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
63
Accuracy:
tensor(0.9625, dtype=torch.float64)
Loss:
tensor(0.0959, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
64
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0441, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
65
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0687, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
66
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0642, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
67
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0448, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
68
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0457, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
69
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0938, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
70
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0431, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
71
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0671, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
72
Accuracy:
tensor(0.9625, dtype=torch.float64)
Loss:
tensor(0.1206, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
73
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0568, grad_fn=<NllLossBackward>)
Epoch:
34
Batch:
74
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0273, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
0
########################
Validation Accuracy:
tensor(0.2933, dtype=torch.float64)
########################
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0670, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
1
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0435, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
2
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0386, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
3
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0592, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
4
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0296, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
5
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0864, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
6
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0345, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
7
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0535, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
8
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0601, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
9
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0326, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
10
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0311, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
11
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0407, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
12
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0415, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
13
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0759, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
14
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0335, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
15
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0528, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
16
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0626, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
17
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0319, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
18
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0525, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
19
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0356, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
20
########################
Validation Accuracy:
tensor(0.2983, dtype=torch.float64)
########################
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0617, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
21
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0562, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
22
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0541, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
23
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0319, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
24
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0504, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
25
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0408, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
26
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0425, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
27
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0301, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
28
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0487, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
29
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0287, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
30
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0217, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
31
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0265, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
32
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0378, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
33
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0446, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
34
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0330, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
35
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0414, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
36
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0329, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
37
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0387, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
38
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0341, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
39
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0377, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
40
########################
Validation Accuracy:
tensor(0.3000, dtype=torch.float64)
########################
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0319, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
41
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0426, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
42
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0318, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
43
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0321, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
44
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0490, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
45
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0470, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
46
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0535, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
47
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0462, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
48
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0510, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
49
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0716, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
50
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0465, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
51
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0586, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
52
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0777, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
53
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0517, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
54
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0652, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
55
Accuracy:
tensor(0.9625, dtype=torch.float64)
Loss:
tensor(0.1027, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
56
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0546, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
57
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0702, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
58
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0803, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
59
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0468, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
60
########################
Validation Accuracy:
tensor(0.3067, dtype=torch.float64)
########################
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0329, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
61
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0488, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
62
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0783, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
63
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0682, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
64
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0222, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
65
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0777, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
66
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0580, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
67
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0453, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
68
Accuracy:
tensor(0.9583, dtype=torch.float64)
Loss:
tensor(0.1086, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
69
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0593, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
70
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0305, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
71
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0376, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
72
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0843, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
73
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0516, grad_fn=<NllLossBackward>)
Epoch:
35
Batch:
74
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0288, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
0
########################
Validation Accuracy:
tensor(0.3050, dtype=torch.float64)
########################
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0209, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
1
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0774, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
2
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0254, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
3
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0243, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
4
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0255, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
5
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0379, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
6
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0254, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
7
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0443, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
8
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0514, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
9
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0374, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
10
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0259, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
11
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0374, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
12
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0211, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
13
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0523, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
14
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0347, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
15
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0360, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
16
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0157, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
17
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0456, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
18
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0480, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
19
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0244, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
20
########################
Validation Accuracy:
tensor(0.3233, dtype=torch.float64)
########################
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0239, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
21
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0512, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
22
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0350, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
23
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0391, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
24
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0420, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
25
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0296, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
26
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0435, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
27
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0732, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
28
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0420, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
29
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0579, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
30
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0325, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
31
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0363, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
32
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0441, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
33
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0535, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
34
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0271, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
35
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0344, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
36
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0684, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
37
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0577, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
38
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0378, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
39
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0277, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
40
########################
Validation Accuracy:
tensor(0.2983, dtype=torch.float64)
########################
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0299, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
41
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0487, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
42
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0392, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
43
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0414, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
44
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0487, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
45
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0629, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
46
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0303, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
47
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0433, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
48
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0378, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
49
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0587, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
50
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0364, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
51
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0365, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
52
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0207, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
53
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0397, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
54
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0445, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
55
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0350, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
56
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0385, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
57
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0420, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
58
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0354, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
59
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0807, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
60
########################
Validation Accuracy:
tensor(0.2917, dtype=torch.float64)
########################
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0640, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
61
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0414, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
62
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0453, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
63
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0548, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
64
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0480, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
65
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0353, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
66
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0359, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
67
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0453, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
68
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0350, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
69
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0263, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
70
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0362, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
71
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0352, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
72
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0370, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
73
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0298, grad_fn=<NllLossBackward>)
Epoch:
36
Batch:
74
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0256, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
0
########################
Validation Accuracy:
tensor(0.2983, dtype=torch.float64)
########################
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0626, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
1
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0203, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
2
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0267, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
3
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0373, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
4
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0382, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
5
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0295, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
6
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0245, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
7
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0201, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
8
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0206, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
9
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0271, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
10
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0213, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
11
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0149, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
12
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0210, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
13
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0415, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
14
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0252, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
15
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0233, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
16
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0327, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
17
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0245, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
18
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0374, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
19
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0414, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
20
########################
Validation Accuracy:
tensor(0.3050, dtype=torch.float64)
########################
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0262, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
21
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0269, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
22
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0287, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
23
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0207, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
24
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0400, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
25
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0435, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
26
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0260, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
27
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0230, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
28
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0373, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
29
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0480, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
30
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0417, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
31
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0314, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
32
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0358, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
33
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0290, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
34
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0194, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
35
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0421, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
36
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0378, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
37
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0281, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
38
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0383, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
39
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0298, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
40
########################
Validation Accuracy:
tensor(0.3000, dtype=torch.float64)
########################
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0392, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
41
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0360, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
42
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0542, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
43
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0631, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
44
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0415, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
45
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0367, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
46
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0347, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
47
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0263, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
48
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0244, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
49
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0403, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
50
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0413, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
51
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0239, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
52
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0463, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
53
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0474, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
54
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0328, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
55
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0400, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
56
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0612, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
57
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0307, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
58
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0573, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
59
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0373, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
60
########################
Validation Accuracy:
tensor(0.3033, dtype=torch.float64)
########################
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0335, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
61
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0277, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
62
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0519, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
63
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0283, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
64
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0414, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
65
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0419, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
66
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0353, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
67
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0392, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
68
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0619, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
69
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0633, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
70
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0571, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
71
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0347, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
72
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0602, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
73
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0353, grad_fn=<NllLossBackward>)
Epoch:
37
Batch:
74
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0323, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
0
########################
Validation Accuracy:
tensor(0.3067, dtype=torch.float64)
########################
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0376, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
1
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0145, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
2
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0369, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
3
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0243, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
4
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0304, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
5
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0213, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
6
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0212, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
7
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0302, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
8
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0447, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
9
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0350, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
10
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0175, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
11
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0451, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
12
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0180, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
13
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0419, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
14
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0576, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
15
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0625, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
16
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0205, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
17
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0477, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
18
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0293, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
19
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0318, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
20
########################
Validation Accuracy:
tensor(0.3000, dtype=torch.float64)
########################
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0353, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
21
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0378, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
22
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0264, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
23
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0404, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
24
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0233, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
25
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0267, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
26
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0339, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
27
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0358, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
28
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0395, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
29
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0246, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
30
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0425, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
31
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0257, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
32
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0465, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
33
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0357, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
34
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0198, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
35
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0358, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
36
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0545, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
37
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0286, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
38
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0283, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
39
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0509, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
40
########################
Validation Accuracy:
tensor(0.3050, dtype=torch.float64)
########################
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0352, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
41
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0364, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
42
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0311, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
43
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0384, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
44
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0339, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
45
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0579, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
46
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0350, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
47
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0355, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
48
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0331, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
49
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0549, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
50
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0315, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
51
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0491, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
52
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0449, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
53
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0221, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
54
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0471, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
55
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0479, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
56
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0413, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
57
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0359, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
58
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0520, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
59
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0461, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
60
########################
Validation Accuracy:
tensor(0.3100, dtype=torch.float64)
########################
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0233, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
61
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0229, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
62
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0395, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
63
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0162, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
64
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0287, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
65
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0366, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
66
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0465, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
67
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0501, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
68
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0339, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
69
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0378, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
70
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0359, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
71
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0555, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
72
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0193, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
73
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0415, grad_fn=<NllLossBackward>)
Epoch:
38
Batch:
74
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0366, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
0
########################
Validation Accuracy:
tensor(0.3017, dtype=torch.float64)
########################
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0188, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
1
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0260, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
2
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0269, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
3
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0586, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
4
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0336, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
5
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0363, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
6
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0362, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
7
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0222, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
8
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0195, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
9
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0571, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
10
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0245, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
11
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0387, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
12
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0326, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
13
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0292, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
14
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0206, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
15
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0204, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
16
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0544, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
17
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0301, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
18
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0160, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
19
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0310, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
20
########################
Validation Accuracy:
tensor(0.3167, dtype=torch.float64)
########################
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0420, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
21
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0304, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
22
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0227, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
23
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0205, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
24
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0468, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
25
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0336, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
26
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0333, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
27
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0307, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
28
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0324, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
29
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0400, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
30
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0579, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
31
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0297, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
32
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0431, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
33
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0327, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
34
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0444, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
35
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0229, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
36
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0357, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
37
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0273, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
38
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0407, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
39
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0366, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
40
########################
Validation Accuracy:
tensor(0.3050, dtype=torch.float64)
########################
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0283, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
41
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0204, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
42
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0451, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
43
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0281, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
44
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0277, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
45
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0347, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
46
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0597, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
47
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0388, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
48
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0308, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
49
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0299, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
50
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0458, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
51
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0303, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
52
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0409, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
53
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0556, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
54
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0442, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
55
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0344, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
56
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0543, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
57
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0512, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
58
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0453, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
59
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0401, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
60
########################
Validation Accuracy:
tensor(0.2883, dtype=torch.float64)
########################
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0424, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
61
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0461, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
62
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0500, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
63
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0460, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
64
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0395, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
65
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0451, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
66
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0214, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
67
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0350, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
68
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0253, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
69
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0281, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
70
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0448, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
71
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0252, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
72
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0268, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
73
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0363, grad_fn=<NllLossBackward>)
Epoch:
39
Batch:
74
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0380, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
0
########################
Validation Accuracy:
tensor(0.3017, dtype=torch.float64)
########################
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0398, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
1
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0313, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
2
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0295, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
3
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0219, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
4
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0273, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
5
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0246, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
6
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0496, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
7
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0187, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
8
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0294, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
9
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0262, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
10
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0394, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
11
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0274, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
12
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0184, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
13
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0506, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
14
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0326, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
15
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0189, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
16
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0506, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
17
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0353, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
18
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0239, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
19
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0222, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
20
########################
Validation Accuracy:
tensor(0.2983, dtype=torch.float64)
########################
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0287, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
21
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0218, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
22
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0344, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
23
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0247, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
24
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0194, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
25
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0443, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
26
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0280, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
27
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0285, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
28
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0302, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
29
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0561, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
30
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0304, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
31
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0551, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
32
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0411, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
33
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0382, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
34
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0408, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
35
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0465, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
36
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0423, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
37
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0436, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
38
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0353, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
39
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0269, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
40
########################
Validation Accuracy:
tensor(0.2900, dtype=torch.float64)
########################
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0205, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
41
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0392, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
42
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0585, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
43
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0274, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
44
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0397, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
45
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0269, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
46
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0244, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
47
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0280, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
48
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0294, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
49
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0224, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
50
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0359, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
51
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0492, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
52
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0251, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
53
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0318, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
54
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0424, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
55
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0298, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
56
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0365, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
57
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0248, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
58
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0355, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
59
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0284, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
60
########################
Validation Accuracy:
tensor(0.2900, dtype=torch.float64)
########################
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0206, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
61
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0216, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
62
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0327, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
63
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0248, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
64
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0490, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
65
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0317, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
66
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0205, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
67
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0411, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
68
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0515, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
69
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0422, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
70
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0308, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
71
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0542, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
72
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0421, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
73
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0279, grad_fn=<NllLossBackward>)
Epoch:
40
Batch:
74
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0447, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
0
########################
Validation Accuracy:
tensor(0.3150, dtype=torch.float64)
########################
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0423, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
1
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0203, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
2
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0406, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
3
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.0738, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
4
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0205, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
5
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0468, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
6
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0291, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
7
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0207, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
8
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0210, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
9
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0308, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
10
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0197, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
11
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0425, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
12
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0306, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
13
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0396, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
14
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0216, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
15
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0339, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
16
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0238, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
17
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0513, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
18
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0323, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
19
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0360, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
20
########################
Validation Accuracy:
tensor(0.3067, dtype=torch.float64)
########################
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0365, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
21
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0426, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
22
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0253, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
23
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0217, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
24
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0361, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
25
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0301, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
26
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0264, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
27
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0339, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
28
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0210, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
29
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0455, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
30
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0301, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
31
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0284, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
32
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0215, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
33
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0345, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
34
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0252, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
35
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0524, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
36
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0194, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
37
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0212, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
38
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0341, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
39
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0260, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
40
########################
Validation Accuracy:
tensor(0.3050, dtype=torch.float64)
########################
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0409, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
41
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0430, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
42
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0200, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
43
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0409, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
44
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0333, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
45
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0273, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
46
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0225, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
47
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0260, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
48
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0265, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
49
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0213, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
50
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0380, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
51
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0201, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
52
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0475, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
53
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0326, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
54
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0248, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
55
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0210, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
56
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0399, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
57
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0492, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
58
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0620, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
59
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0503, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
60
########################
Validation Accuracy:
tensor(0.3150, dtype=torch.float64)
########################
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0422, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
61
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0276, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
62
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0226, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
63
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0533, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
64
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0464, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
65
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0260, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
66
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0451, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
67
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0498, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
68
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0496, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
69
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0567, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
70
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0296, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
71
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0328, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
72
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0406, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
73
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0292, grad_fn=<NllLossBackward>)
Epoch:
41
Batch:
74
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0522, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
0
########################
Validation Accuracy:
tensor(0.2950, dtype=torch.float64)
########################
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0438, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
1
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0257, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
2
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0231, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
3
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0319, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
4
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0330, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
5
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0227, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
6
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0356, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
7
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0331, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
8
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0400, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
9
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0123, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
10
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0363, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
11
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0486, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
12
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0263, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
13
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0273, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
14
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0266, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
15
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0274, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
16
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0313, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
17
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0247, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
18
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0141, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
19
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0510, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
20
########################
Validation Accuracy:
tensor(0.2950, dtype=torch.float64)
########################
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0328, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
21
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0348, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
22
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0371, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
23
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0300, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
24
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0446, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
25
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0185, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
26
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0176, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
27
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0312, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
28
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0217, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
29
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0319, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
30
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0424, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
31
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0142, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
32
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0204, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
33
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0351, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
34
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0344, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
35
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0299, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
36
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0204, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
37
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0396, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
38
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0404, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
39
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0262, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
40
########################
Validation Accuracy:
tensor(0.2967, dtype=torch.float64)
########################
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0223, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
41
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0307, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
42
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0214, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
43
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0310, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
44
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0340, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
45
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0653, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
46
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0531, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
47
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0366, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
48
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0405, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
49
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0445, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
50
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0224, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
51
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0466, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
52
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0302, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
53
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0262, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
54
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0486, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
55
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0519, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
56
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0478, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
57
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0332, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
58
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0310, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
59
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0266, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
60
########################
Validation Accuracy:
tensor(0.3000, dtype=torch.float64)
########################
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0144, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
61
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0248, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
62
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0480, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
63
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0258, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
64
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0331, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
65
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0359, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
66
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0228, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
67
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0441, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
68
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0505, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
69
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0336, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
70
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0492, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
71
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0208, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
72
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0253, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
73
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0238, grad_fn=<NllLossBackward>)
Epoch:
42
Batch:
74
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0153, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
0
########################
Validation Accuracy:
tensor(0.3100, dtype=torch.float64)
########################
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0427, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
1
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0173, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
2
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0191, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
3
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0245, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
4
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0205, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
5
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0308, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
6
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0254, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
7
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0277, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
8
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0173, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
9
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0383, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
10
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0198, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
11
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0186, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
12
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0229, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
13
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0225, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
14
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0232, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
15
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0401, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
16
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0261, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
17
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0318, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
18
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0228, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
19
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0539, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
20
########################
Validation Accuracy:
tensor(0.3050, dtype=torch.float64)
########################
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0438, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
21
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0219, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
22
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0295, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
23
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0309, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
24
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0177, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
25
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0166, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
26
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0387, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
27
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0297, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
28
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0378, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
29
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0197, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
30
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0251, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
31
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0264, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
32
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0375, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
33
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0271, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
34
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0224, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
35
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0342, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
36
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0442, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
37
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0155, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
38
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0282, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
39
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0471, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
40
########################
Validation Accuracy:
tensor(0.3067, dtype=torch.float64)
########################
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0276, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
41
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0295, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
42
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0200, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
43
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0298, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
44
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0306, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
45
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0229, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
46
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0493, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
47
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0346, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
48
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0427, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
49
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0419, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
50
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0392, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
51
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0286, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
52
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0213, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
53
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0290, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
54
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0467, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
55
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0358, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
56
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0230, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
57
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0254, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
58
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0585, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
59
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0236, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
60
########################
Validation Accuracy:
tensor(0.3167, dtype=torch.float64)
########################
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0362, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
61
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0436, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
62
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0192, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
63
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0242, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
64
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0394, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
65
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0203, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
66
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0449, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
67
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0590, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
68
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0362, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
69
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0331, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
70
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0447, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
71
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0294, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
72
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0769, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
73
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0676, grad_fn=<NllLossBackward>)
Epoch:
43
Batch:
74
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0198, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
0
########################
Validation Accuracy:
tensor(0.3117, dtype=torch.float64)
########################
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0228, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
1
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0218, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
2
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0355, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
3
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0375, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
4
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0363, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
5
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0345, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
6
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0284, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
7
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0306, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
8
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0413, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
9
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0448, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
10
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0235, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
11
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0407, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
12
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0218, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
13
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0543, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
14
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0311, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
15
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0290, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
16
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0275, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
17
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0405, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
18
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0542, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
19
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0205, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
20
########################
Validation Accuracy:
tensor(0.2867, dtype=torch.float64)
########################
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0318, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
21
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0527, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
22
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0257, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
23
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0242, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
24
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0290, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
25
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0329, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
26
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0550, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
27
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0115, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
28
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0435, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
29
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0305, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
30
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0157, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
31
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0311, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
32
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0321, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
33
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0213, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
34
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0187, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
35
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0431, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
36
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0435, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
37
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0373, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
38
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0311, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
39
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0242, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
40
########################
Validation Accuracy:
tensor(0.3217, dtype=torch.float64)
########################
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0161, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
41
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0430, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
42
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0304, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
43
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0295, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
44
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0336, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
45
Accuracy:
tensor(0.9500, dtype=torch.float64)
Loss:
tensor(0.1320, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
46
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0288, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
47
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0445, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
48
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0745, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
49
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0365, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
50
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0261, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
51
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0754, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
52
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0694, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
53
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0535, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
54
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0223, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
55
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0418, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
56
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0710, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
57
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0344, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
58
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0785, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
59
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0535, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
60
########################
Validation Accuracy:
tensor(0.2833, dtype=torch.float64)
########################
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0530, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
61
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0410, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
62
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0372, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
63
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0322, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
64
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0426, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
65
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0330, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
66
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0566, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
67
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0314, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
68
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0363, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
69
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0357, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
70
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0388, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
71
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0316, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
72
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0356, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
73
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0277, grad_fn=<NllLossBackward>)
Epoch:
44
Batch:
74
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0407, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
0
########################
Validation Accuracy:
tensor(0.3033, dtype=torch.float64)
########################
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0406, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
1
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0650, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
2
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0244, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
3
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0323, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
4
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0222, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
5
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0476, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
6
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0316, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
7
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0448, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
8
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0333, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
9
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0293, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
10
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0316, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
11
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0356, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
12
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0325, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
13
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0219, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
14
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0155, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
15
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0334, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
16
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0265, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
17
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0461, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
18
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0282, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
19
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0220, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
20
########################
Validation Accuracy:
tensor(0.2983, dtype=torch.float64)
########################
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0408, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
21
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0218, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
22
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0326, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
23
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0451, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
24
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0301, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
25
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0164, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
26
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0407, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
27
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0306, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
28
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0365, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
29
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0496, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
30
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0412, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
31
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0507, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
32
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0320, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
33
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0515, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
34
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0345, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
35
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0806, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
36
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0302, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
37
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0255, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
38
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0215, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
39
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0286, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
40
########################
Validation Accuracy:
tensor(0.2983, dtype=torch.float64)
########################
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0286, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
41
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0259, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
42
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0316, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
43
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0326, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
44
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0672, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
45
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0558, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
46
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0299, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
47
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0265, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
48
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0218, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
49
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0265, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
50
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0399, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
51
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0557, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
52
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0379, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
53
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0257, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
54
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0335, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
55
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0414, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
56
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0263, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
57
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0294, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
58
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0390, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
59
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0494, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
60
########################
Validation Accuracy:
tensor(0.2900, dtype=torch.float64)
########################
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0408, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
61
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0330, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
62
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0251, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
63
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0466, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
64
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0518, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
65
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0206, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
66
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0396, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
67
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0683, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
68
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0600, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
69
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0222, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
70
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0366, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
71
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.0943, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
72
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0417, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
73
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0648, grad_fn=<NllLossBackward>)
Epoch:
45
Batch:
74
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0715, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
0
########################
Validation Accuracy:
tensor(0.2917, dtype=torch.float64)
########################
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0327, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
1
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0670, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
2
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0741, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
3
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0832, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
4
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0316, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
5
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0446, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
6
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0884, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
7
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0458, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
8
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0266, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
9
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0290, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
10
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0247, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
11
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0492, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
12
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0619, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
13
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0207, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
14
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0328, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
15
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0489, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
16
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0238, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
17
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0251, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
18
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0245, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
19
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0351, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
20
########################
Validation Accuracy:
tensor(0.2933, dtype=torch.float64)
########################
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0275, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
21
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0482, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
22
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0413, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
23
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0325, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
24
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0414, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
25
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0288, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
26
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0423, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
27
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0347, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
28
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0311, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
29
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0360, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
30
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0443, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
31
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0420, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
32
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0589, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
33
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0252, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
34
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0287, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
35
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0616, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
36
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0273, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
37
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0312, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
38
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0406, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
39
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0610, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
40
########################
Validation Accuracy:
tensor(0.2967, dtype=torch.float64)
########################
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0454, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
41
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0336, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
42
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0361, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
43
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0509, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
44
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0448, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
45
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0379, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
46
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0408, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
47
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0322, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
48
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0386, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
49
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0514, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
50
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0226, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
51
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0577, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
52
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0400, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
53
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0329, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
54
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0709, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
55
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0420, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
56
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0655, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
57
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0688, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
58
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0353, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
59
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0360, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
60
########################
Validation Accuracy:
tensor(0.2850, dtype=torch.float64)
########################
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0608, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
61
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0706, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
62
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0351, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
63
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0412, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
64
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0416, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
65
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0362, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
66
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0426, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
67
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0679, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
68
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0684, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
69
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0289, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
70
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0503, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
71
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0385, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
72
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0560, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
73
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0586, grad_fn=<NllLossBackward>)
Epoch:
46
Batch:
74
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0237, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
0
########################
Validation Accuracy:
tensor(0.3017, dtype=torch.float64)
########################
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0185, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
1
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0245, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
2
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0344, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
3
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0444, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
4
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0393, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
5
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0214, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
6
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0438, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
7
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0481, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
8
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0205, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
9
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0318, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
10
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0271, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
11
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0384, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
12
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0424, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
13
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0413, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
14
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0295, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
15
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0361, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
16
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0349, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
17
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0207, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
18
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0254, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
19
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0494, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
20
########################
Validation Accuracy:
tensor(0.2883, dtype=torch.float64)
########################
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0378, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
21
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0246, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
22
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0501, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
23
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0296, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
24
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0415, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
25
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0489, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
26
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0273, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
27
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0251, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
28
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0278, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
29
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0164, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
30
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0716, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
31
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0320, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
32
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0343, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
33
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0342, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
34
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0256, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
35
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0302, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
36
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.1039, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
37
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0498, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
38
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0277, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
39
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0480, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
40
########################
Validation Accuracy:
tensor(0.2933, dtype=torch.float64)
########################
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0323, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
41
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0300, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
42
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0954, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
43
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0308, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
44
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0390, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
45
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0422, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
46
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0197, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
47
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0220, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
48
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0492, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
49
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0199, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
50
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0653, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
51
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0510, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
52
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0537, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
53
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0513, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
54
Accuracy:
tensor(0.9542, dtype=torch.float64)
Loss:
tensor(0.1231, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
55
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0300, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
56
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0641, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
57
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0481, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
58
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0654, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
59
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0660, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
60
########################
Validation Accuracy:
tensor(0.2850, dtype=torch.float64)
########################
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0423, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
61
Accuracy:
tensor(0.9625, dtype=torch.float64)
Loss:
tensor(0.1096, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
62
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0506, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
63
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0548, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
64
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0493, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
65
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0456, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
66
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0566, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
67
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0146, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
68
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0438, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
69
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0288, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
70
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0187, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
71
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0943, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
72
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0801, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
73
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0331, grad_fn=<NllLossBackward>)
Epoch:
47
Batch:
74
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0276, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
0
########################
Validation Accuracy:
tensor(0.2967, dtype=torch.float64)
########################
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0217, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
1
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0259, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
2
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0584, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
3
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0173, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
4
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0450, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
5
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0330, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
6
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0218, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
7
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0305, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
8
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0165, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
9
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0299, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
10
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0238, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
11
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0298, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
12
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0340, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
13
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0366, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
14
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0436, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
15
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0357, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
16
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0116, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
17
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0341, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
18
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0266, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
19
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0264, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
20
########################
Validation Accuracy:
tensor(0.2900, dtype=torch.float64)
########################
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0251, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
21
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0264, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
22
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0389, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
23
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0178, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
24
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0390, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
25
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0439, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
26
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0259, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
27
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0280, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
28
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0209, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
29
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0314, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
30
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0127, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
31
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0265, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
32
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0185, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
33
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0454, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
34
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0285, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
35
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0208, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
36
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0332, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
37
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0377, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
38
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0285, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
39
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0206, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
40
########################
Validation Accuracy:
tensor(0.2883, dtype=torch.float64)
########################
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0227, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
41
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0417, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
42
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0550, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
43
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0368, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
44
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0457, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
45
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0265, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
46
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0584, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
47
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0265, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
48
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0191, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
49
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0153, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
50
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0279, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
51
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0158, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
52
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0401, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
53
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0303, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
54
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0229, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
55
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0176, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
56
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0314, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
57
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0430, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
58
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0288, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
59
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0301, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
60
########################
Validation Accuracy:
tensor(0.2917, dtype=torch.float64)
########################
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0195, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
61
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0321, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
62
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0234, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
63
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0635, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
64
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0267, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
65
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0900, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
66
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0450, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
67
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0579, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
68
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0619, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
69
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0605, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
70
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0198, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
71
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0291, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
72
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0393, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
73
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0197, grad_fn=<NllLossBackward>)
Epoch:
48
Batch:
74
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0644, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
0
########################
Validation Accuracy:
tensor(0.3017, dtype=torch.float64)
########################
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0517, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
1
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0367, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
2
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0237, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
3
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0466, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
4
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0159, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
5
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0591, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
6
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0338, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
7
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0484, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
8
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0517, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
9
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0301, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
10
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0321, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
11
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0559, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
12
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0344, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
13
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0561, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
14
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0288, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
15
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0330, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
16
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0142, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
17
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0460, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
18
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0258, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
19
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0418, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
20
########################
Validation Accuracy:
tensor(0.2933, dtype=torch.float64)
########################
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0348, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
21
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0465, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
22
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0355, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
23
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0220, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
24
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0309, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
25
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0306, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
26
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0299, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
27
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0506, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
28
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0113, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
29
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0196, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
30
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0236, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
31
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0220, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
32
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0366, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
33
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0246, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
34
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0343, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
35
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0246, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
36
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0239, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
37
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0273, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
38
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0323, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
39
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0311, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
40
########################
Validation Accuracy:
tensor(0.3033, dtype=torch.float64)
########################
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0358, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
41
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0179, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
42
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0289, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
43
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0098, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
44
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0373, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
45
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0223, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
46
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0200, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
47
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0122, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
48
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0234, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
49
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0219, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
50
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0415, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
51
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0358, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
52
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0437, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
53
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0245, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
54
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0296, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
55
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0197, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
56
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0512, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
57
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0527, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
58
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0196, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
59
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0312, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
60
########################
Validation Accuracy:
tensor(0.3117, dtype=torch.float64)
########################
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0279, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
61
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0430, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
62
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0377, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
63
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0230, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
64
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0243, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
65
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0193, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
66
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0461, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
67
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0424, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
68
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0199, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
69
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0308, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
70
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0297, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
71
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0407, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
72
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0329, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
73
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0244, grad_fn=<NllLossBackward>)
Epoch:
49
Batch:
74
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0405, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
0
########################
Validation Accuracy:
tensor(0.3217, dtype=torch.float64)
########################
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0387, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
1
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0299, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
2
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0197, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
3
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0470, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
4
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0284, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
5
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0197, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
6
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0177, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
7
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0151, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
8
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0140, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
9
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0234, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
10
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0398, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
11
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0440, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
12
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0294, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
13
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0370, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
14
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0193, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
15
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0271, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
16
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0339, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
17
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0334, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
18
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0326, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
19
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0374, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
20
########################
Validation Accuracy:
tensor(0.2917, dtype=torch.float64)
########################
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0212, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
21
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0236, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
22
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0224, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
23
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0294, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
24
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0197, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
25
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0432, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
26
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0241, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
27
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0248, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
28
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0220, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
29
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0523, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
30
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0160, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
31
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0199, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
32
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0259, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
33
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0228, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
34
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0130, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
35
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0320, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
36
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0135, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
37
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0308, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
38
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0201, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
39
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0293, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
40
########################
Validation Accuracy:
tensor(0.3000, dtype=torch.float64)
########################
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0215, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
41
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0286, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
42
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0368, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
43
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0318, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
44
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0258, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
45
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0383, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
46
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0181, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
47
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0381, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
48
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0171, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
49
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0264, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
50
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0426, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
51
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0365, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
52
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0226, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
53
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0185, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
54
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0357, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
55
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0636, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
56
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0201, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
57
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0315, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
58
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0972, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
59
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0193, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
60
########################
Validation Accuracy:
tensor(0.3033, dtype=torch.float64)
########################
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0397, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
61
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0644, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
62
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0385, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
63
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0396, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
64
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0334, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
65
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0484, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
66
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0394, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
67
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0397, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
68
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0431, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
69
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0399, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
70
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0425, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
71
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0350, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
72
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0329, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
73
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0267, grad_fn=<NllLossBackward>)
Epoch:
50
Batch:
74
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0227, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
0
########################
Validation Accuracy:
tensor(0.2967, dtype=torch.float64)
########################
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0295, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
1
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0330, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
2
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0104, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
3
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0112, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
4
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0281, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
5
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0088, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
6
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0199, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
7
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0258, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
8
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0205, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
9
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0257, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
10
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0358, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
11
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0171, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
12
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0356, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
13
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0206, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
14
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0170, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
15
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0269, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
16
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0362, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
17
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0225, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
18
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0146, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
19
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0316, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
20
########################
Validation Accuracy:
tensor(0.2967, dtype=torch.float64)
########################
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0382, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
21
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0203, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
22
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0127, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
23
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0198, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
24
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0210, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
25
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0279, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
26
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0249, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
27
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0286, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
28
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0176, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
29
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0210, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
30
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0209, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
31
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0237, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
32
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0192, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
33
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0297, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
34
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0251, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
35
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0462, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
36
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0400, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
37
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0227, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
38
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0169, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
39
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0189, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
40
########################
Validation Accuracy:
tensor(0.2950, dtype=torch.float64)
########################
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0285, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
41
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0201, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
42
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0372, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
43
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0230, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
44
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0303, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
45
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0222, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
46
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0432, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
47
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0196, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
48
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0395, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
49
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0260, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
50
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0206, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
51
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0230, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
52
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0105, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
53
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0178, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
54
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0286, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
55
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0319, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
56
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0287, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
57
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0513, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
58
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0270, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
59
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0132, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
60
########################
Validation Accuracy:
tensor(0.3050, dtype=torch.float64)
########################
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0255, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
61
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0295, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
62
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0611, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
63
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0206, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
64
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0173, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
65
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0468, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
66
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0235, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
67
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0218, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
68
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0339, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
69
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0450, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
70
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0175, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
71
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0189, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
72
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0272, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
73
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0572, grad_fn=<NllLossBackward>)
Epoch:
51
Batch:
74
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0138, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
0
########################
Validation Accuracy:
tensor(0.2983, dtype=torch.float64)
########################
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0204, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
1
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0292, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
2
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0206, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
3
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0107, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
4
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0250, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
5
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0287, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
6
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0236, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
7
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0323, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
8
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0171, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
9
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0349, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
10
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0312, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
11
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0388, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
12
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0161, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
13
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0131, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
14
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0278, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
15
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0180, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
16
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0108, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
17
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0226, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
18
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0310, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
19
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0290, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
20
########################
Validation Accuracy:
tensor(0.3017, dtype=torch.float64)
########################
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0214, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
21
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0318, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
22
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0326, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
23
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0234, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
24
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0263, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
25
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0129, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
26
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0276, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
27
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0180, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
28
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0308, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
29
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0337, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
30
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0261, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
31
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0183, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
32
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0490, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
33
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0272, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
34
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0187, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
35
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0407, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
36
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0356, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
37
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0159, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
38
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0567, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
39
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0159, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
40
########################
Validation Accuracy:
tensor(0.2883, dtype=torch.float64)
########################
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0321, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
41
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0516, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
42
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0408, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
43
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0435, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
44
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0621, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
45
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0350, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
46
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0183, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
47
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0346, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
48
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0589, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
49
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0205, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
50
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0392, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
51
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0416, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
52
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0364, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
53
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0207, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
54
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0710, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
55
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0149, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
56
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0285, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
57
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0889, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
58
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0279, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
59
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0213, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
60
########################
Validation Accuracy:
tensor(0.3033, dtype=torch.float64)
########################
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0199, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
61
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0338, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
62
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0195, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
63
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0251, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
64
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0455, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
65
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0334, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
66
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0403, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
67
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0195, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
68
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0322, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
69
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0602, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
70
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0282, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
71
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0213, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
72
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0178, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
73
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0393, grad_fn=<NllLossBackward>)
Epoch:
52
Batch:
74
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0158, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
0
########################
Validation Accuracy:
tensor(0.3017, dtype=torch.float64)
########################
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0351, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
1
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0277, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
2
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0379, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
3
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0253, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
4
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0401, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
5
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0511, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
6
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0097, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
7
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0150, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
8
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0182, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
9
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0258, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
10
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0287, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
11
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0289, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
12
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0204, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
13
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0235, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
14
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0227, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
15
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0477, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
16
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0314, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
17
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0293, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
18
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0115, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
19
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0234, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
20
########################
Validation Accuracy:
tensor(0.3100, dtype=torch.float64)
########################
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0304, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
21
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0463, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
22
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0340, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
23
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0161, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
24
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0277, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
25
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0295, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
26
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0188, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
27
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0263, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
28
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0526, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
29
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0249, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
30
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0181, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
31
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0539, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
32
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0227, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
33
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0360, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
34
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0171, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
35
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0434, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
36
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0328, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
37
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0283, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
38
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0178, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
39
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0313, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
40
########################
Validation Accuracy:
tensor(0.3017, dtype=torch.float64)
########################
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0265, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
41
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0236, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
42
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0498, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
43
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0320, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
44
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0331, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
45
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0240, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
46
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0394, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
47
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0204, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
48
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0329, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
49
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0214, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
50
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0693, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
51
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0343, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
52
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0269, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
53
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0312, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
54
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0236, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
55
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0209, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
56
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0491, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
57
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0487, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
58
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0131, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
59
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0512, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
60
########################
Validation Accuracy:
tensor(0.2817, dtype=torch.float64)
########################
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0573, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
61
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0268, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
62
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0515, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
63
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0455, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
64
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0272, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
65
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0245, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
66
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0632, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
67
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0328, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
68
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0558, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
69
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0290, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
70
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0419, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
71
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0274, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
72
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0332, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
73
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0203, grad_fn=<NllLossBackward>)
Epoch:
53
Batch:
74
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0443, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
0
########################
Validation Accuracy:
tensor(0.3000, dtype=torch.float64)
########################
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0468, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
1
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0160, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
2
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0466, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
3
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0305, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
4
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0569, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
5
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0435, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
6
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0219, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
7
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0168, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
8
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0330, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
9
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0195, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
10
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0528, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
11
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0248, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
12
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0379, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
13
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0504, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
14
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0257, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
15
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0192, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
16
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0194, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
17
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0514, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
18
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0267, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
19
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0291, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
20
########################
Validation Accuracy:
tensor(0.2983, dtype=torch.float64)
########################
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0382, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
21
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0217, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
22
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0512, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
23
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0364, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
24
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0186, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
25
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0115, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
26
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0262, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
27
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0540, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
28
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0241, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
29
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0149, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
30
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0457, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
31
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0185, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
32
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0198, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
33
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0260, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
34
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0931, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
35
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0318, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
36
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0844, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
37
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0437, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
38
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0184, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
39
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0346, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
40
########################
Validation Accuracy:
tensor(0.2933, dtype=torch.float64)
########################
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0258, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
41
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0329, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
42
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0406, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
43
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0289, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
44
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0597, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
45
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0454, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
46
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0498, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
47
Accuracy:
tensor(0.9625, dtype=torch.float64)
Loss:
tensor(0.0864, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
48
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0699, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
49
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0519, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
50
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0252, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
51
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0507, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
52
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0480, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
53
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0347, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
54
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0286, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
55
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0641, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
56
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0199, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
57
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0201, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
58
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0555, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
59
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0474, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
60
########################
Validation Accuracy:
tensor(0.2950, dtype=torch.float64)
########################
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0190, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
61
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0139, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
62
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0363, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
63
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0348, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
64
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0227, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
65
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0208, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
66
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0553, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
67
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0318, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
68
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0431, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
69
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0567, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
70
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0230, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
71
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0214, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
72
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0168, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
73
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0329, grad_fn=<NllLossBackward>)
Epoch:
54
Batch:
74
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0406, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
0
########################
Validation Accuracy:
tensor(0.2900, dtype=torch.float64)
########################
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0235, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
1
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0483, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
2
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0281, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
3
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0211, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
4
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0243, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
5
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0216, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
6
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0364, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
7
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0401, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
8
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0211, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
9
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0371, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
10
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0376, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
11
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0625, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
12
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0294, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
13
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0320, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
14
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0615, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
15
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0286, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
16
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0122, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
17
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0249, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
18
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0306, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
19
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0341, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
20
########################
Validation Accuracy:
tensor(0.3000, dtype=torch.float64)
########################
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0196, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
21
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0497, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
22
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0543, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
23
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0421, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
24
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0205, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
25
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0250, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
26
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.1030, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
27
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0330, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
28
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0103, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
29
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0330, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
30
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0422, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
31
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0513, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
32
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0272, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
33
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0217, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
34
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0429, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
35
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0255, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
36
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0140, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
37
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0249, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
38
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0278, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
39
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0317, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
40
########################
Validation Accuracy:
tensor(0.2917, dtype=torch.float64)
########################
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0258, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
41
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0197, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
42
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0720, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
43
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0407, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
44
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0149, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
45
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0233, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
46
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0312, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
47
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0231, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
48
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0206, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
49
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0240, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
50
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0204, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
51
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0333, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
52
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0195, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
53
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0336, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
54
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0234, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
55
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0513, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
56
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0530, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
57
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0276, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
58
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0556, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
59
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0317, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
60
########################
Validation Accuracy:
tensor(0.2933, dtype=torch.float64)
########################
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0231, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
61
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0228, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
62
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0365, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
63
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0105, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
64
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0307, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
65
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0391, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
66
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0595, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
67
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0537, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
68
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0461, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
69
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0387, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
70
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0610, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
71
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0376, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
72
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0274, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
73
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0436, grad_fn=<NllLossBackward>)
Epoch:
55
Batch:
74
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0276, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
0
########################
Validation Accuracy:
tensor(0.2917, dtype=torch.float64)
########################
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0372, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
1
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0427, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
2
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0424, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
3
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0146, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
4
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0310, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
5
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0307, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
6
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0769, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
7
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0338, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
8
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0883, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
9
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0149, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
10
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0254, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
11
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0403, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
12
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0573, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
13
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0186, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
14
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0307, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
15
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0623, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
16
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0528, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
17
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0197, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
18
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0216, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
19
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0258, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
20
########################
Validation Accuracy:
tensor(0.2933, dtype=torch.float64)
########################
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0590, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
21
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0275, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
22
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0443, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
23
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0281, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
24
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0201, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
25
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0340, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
26
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0689, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
27
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0169, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
28
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0227, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
29
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0316, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
30
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0847, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
31
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0150, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
32
Accuracy:
tensor(0.9542, dtype=torch.float64)
Loss:
tensor(0.0960, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
33
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0358, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
34
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0566, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
35
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0361, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
36
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0182, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
37
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0164, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
38
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0150, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
39
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0458, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
40
########################
Validation Accuracy:
tensor(0.2950, dtype=torch.float64)
########################
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0541, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
41
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0284, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
42
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0110, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
43
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0499, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
44
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0340, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
45
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0270, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
46
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0299, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
47
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0602, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
48
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0393, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
49
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0283, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
50
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0268, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
51
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0299, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
52
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0410, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
53
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0227, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
54
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0310, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
55
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0541, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
56
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0555, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
57
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0673, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
58
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0314, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
59
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0187, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
60
########################
Validation Accuracy:
tensor(0.3117, dtype=torch.float64)
########################
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0358, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
61
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0298, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
62
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0346, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
63
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0292, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
64
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0163, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
65
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0563, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
66
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0394, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
67
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0375, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
68
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0182, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
69
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0299, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
70
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0376, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
71
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0344, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
72
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0245, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
73
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0306, grad_fn=<NllLossBackward>)
Epoch:
56
Batch:
74
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0132, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
0
########################
Validation Accuracy:
tensor(0.2917, dtype=torch.float64)
########################
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0326, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
1
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0404, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
2
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0148, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
3
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0229, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
4
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0357, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
5
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0241, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
6
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0514, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
7
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0461, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
8
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0079, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
9
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0214, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
10
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0483, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
11
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0165, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
12
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0366, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
13
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0174, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
14
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0183, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
15
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0407, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
16
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0167, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
17
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0202, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
18
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0252, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
19
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0185, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
20
########################
Validation Accuracy:
tensor(0.2900, dtype=torch.float64)
########################
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0286, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
21
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0214, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
22
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0566, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
23
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0156, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
24
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0203, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
25
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0270, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
26
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0155, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
27
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0181, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
28
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0497, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
29
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0187, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
30
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0381, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
31
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0271, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
32
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0302, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
33
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0312, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
34
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0282, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
35
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0481, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
36
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0130, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
37
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0429, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
38
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0252, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
39
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0146, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
40
########################
Validation Accuracy:
tensor(0.3150, dtype=torch.float64)
########################
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0415, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
41
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0209, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
42
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0268, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
43
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0222, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
44
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0321, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
45
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0274, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
46
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0275, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
47
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0402, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
48
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0163, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
49
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0435, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
50
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0191, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
51
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0356, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
52
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0264, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
53
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0333, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
54
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0654, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
55
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0477, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
56
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0259, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
57
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0276, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
58
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0248, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
59
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0313, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
60
########################
Validation Accuracy:
tensor(0.3033, dtype=torch.float64)
########################
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0448, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
61
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0355, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
62
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0278, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
63
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0310, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
64
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0431, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
65
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0599, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
66
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0115, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
67
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0538, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
68
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0237, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
69
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0531, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
70
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0262, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
71
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0194, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
72
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0329, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
73
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0195, grad_fn=<NllLossBackward>)
Epoch:
57
Batch:
74
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0283, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
0
########################
Validation Accuracy:
tensor(0.2950, dtype=torch.float64)
########################
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0220, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
1
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0354, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
2
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0197, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
3
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0289, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
4
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0725, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
5
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0310, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
6
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0323, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
7
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0328, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
8
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0440, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
9
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0513, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
10
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0589, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
11
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0145, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
12
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0259, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
13
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0416, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
14
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0106, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
15
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0269, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
16
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0478, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
17
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0281, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
18
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0157, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
19
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0134, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
20
########################
Validation Accuracy:
tensor(0.3000, dtype=torch.float64)
########################
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0504, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
21
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0136, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
22
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0179, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
23
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0131, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
24
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0207, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
25
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0255, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
26
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0367, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
27
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0194, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
28
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0285, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
29
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0211, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
30
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0237, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
31
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0267, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
32
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0283, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
33
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0180, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
34
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0205, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
35
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0273, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
36
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0274, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
37
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0511, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
38
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0297, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
39
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0198, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
40
########################
Validation Accuracy:
tensor(0.2917, dtype=torch.float64)
########################
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0387, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
41
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0321, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
42
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0162, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
43
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0184, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
44
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0290, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
45
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0200, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
46
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0188, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
47
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0311, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
48
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0225, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
49
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0394, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
50
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0254, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
51
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0256, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
52
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0404, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
53
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0432, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
54
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0448, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
55
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0210, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
56
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0223, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
57
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0252, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
58
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0345, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
59
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0576, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
60
########################
Validation Accuracy:
tensor(0.3033, dtype=torch.float64)
########################
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0385, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
61
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0210, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
62
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0301, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
63
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0344, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
64
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0229, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
65
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0300, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
66
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0307, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
67
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0207, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
68
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0259, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
69
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0234, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
70
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0191, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
71
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0395, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
72
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0411, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
73
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0146, grad_fn=<NllLossBackward>)
Epoch:
58
Batch:
74
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0270, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
0
########################
Validation Accuracy:
tensor(0.2983, dtype=torch.float64)
########################
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0508, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
1
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0195, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
2
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0329, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
3
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0164, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
4
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0244, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
5
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0145, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
6
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0207, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
7
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0154, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
8
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0120, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
9
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0361, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
10
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0294, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
11
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0214, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
12
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0133, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
13
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0219, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
14
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0302, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
15
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0358, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
16
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0407, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
17
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0238, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
18
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0212, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
19
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0454, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
20
########################
Validation Accuracy:
tensor(0.2800, dtype=torch.float64)
########################
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0325, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
21
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0205, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
22
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0310, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
23
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0176, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
24
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0431, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
25
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0149, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
26
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0381, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
27
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0205, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
28
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0405, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
29
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0165, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
30
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0250, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
31
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0174, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
32
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0430, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
33
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0384, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
34
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0159, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
35
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0597, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
36
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0225, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
37
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0215, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
38
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0339, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
39
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0304, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
40
########################
Validation Accuracy:
tensor(0.3083, dtype=torch.float64)
########################
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0253, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
41
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0540, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
42
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0127, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
43
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0242, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
44
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0233, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
45
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0443, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
46
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0270, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
47
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0324, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
48
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0245, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
49
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0116, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
50
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0345, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
51
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0200, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
52
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0360, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
53
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0369, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
54
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0272, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
55
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0249, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
56
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0420, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
57
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0279, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
58
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0233, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
59
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0240, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
60
########################
Validation Accuracy:
tensor(0.3083, dtype=torch.float64)
########################
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0373, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
61
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0678, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
62
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0125, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
63
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0124, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
64
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0281, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
65
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0520, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
66
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0543, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
67
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0249, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
68
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0592, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
69
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0553, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
70
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0308, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
71
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0170, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
72
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0256, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
73
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0193, grad_fn=<NllLossBackward>)
Epoch:
59
Batch:
74
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0376, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
0
########################
Validation Accuracy:
tensor(0.2750, dtype=torch.float64)
########################
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0429, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
1
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0324, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
2
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0187, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
3
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0184, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
4
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0104, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
5
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0178, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
6
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0286, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
7
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0163, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
8
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0081, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
9
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0327, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
10
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0297, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
11
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0360, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
12
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0117, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
13
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0325, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
14
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0341, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
15
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0387, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
16
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0194, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
17
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0292, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
18
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0172, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
19
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0136, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
20
########################
Validation Accuracy:
tensor(0.2950, dtype=torch.float64)
########################
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0182, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
21
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0255, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
22
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0115, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
23
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0538, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
24
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0157, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
25
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0419, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
26
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0433, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
27
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0304, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
28
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0130, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
29
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0554, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
30
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0338, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
31
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0151, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
32
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0268, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
33
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0376, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
34
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0526, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
35
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0346, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
36
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0258, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
37
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0289, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
38
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0265, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
39
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0153, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
40
########################
Validation Accuracy:
tensor(0.2850, dtype=torch.float64)
########################
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0144, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
41
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0301, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
42
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0517, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
43
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0195, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
44
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0179, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
45
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0129, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
46
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0399, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
47
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0308, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
48
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0286, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
49
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0189, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
50
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0347, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
51
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0233, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
52
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0210, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
53
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0643, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
54
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0404, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
55
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0353, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
56
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0302, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
57
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0291, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
58
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0158, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
59
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0174, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
60
########################
Validation Accuracy:
tensor(0.2983, dtype=torch.float64)
########################
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0442, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
61
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0206, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
62
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0211, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
63
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0455, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
64
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0182, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
65
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0179, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
66
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0373, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
67
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0863, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
68
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0405, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
69
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0330, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
70
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0179, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
71
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0110, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
72
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0466, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
73
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0230, grad_fn=<NllLossBackward>)
Epoch:
60
Batch:
74
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0294, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
0
########################
Validation Accuracy:
tensor(0.2950, dtype=torch.float64)
########################
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0213, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
1
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0209, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
2
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0433, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
3
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0179, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
4
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0184, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
5
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0242, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
6
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0574, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
7
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0098, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
8
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0377, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
9
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0310, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
10
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0095, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
11
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0117, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
12
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0335, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
13
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0442, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
14
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0174, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
15
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0113, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
16
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0177, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
17
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0216, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
18
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0089, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
19
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0107, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
20
########################
Validation Accuracy:
tensor(0.2967, dtype=torch.float64)
########################
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0439, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
21
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0314, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
22
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0192, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
23
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0184, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
24
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0280, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
25
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0362, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
26
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0420, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
27
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0221, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
28
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0394, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
29
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0113, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
30
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0265, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
31
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0324, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
32
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0344, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
33
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0223, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
34
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0434, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
35
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0132, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
36
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0267, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
37
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0122, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
38
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0506, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
39
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0271, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
40
########################
Validation Accuracy:
tensor(0.2950, dtype=torch.float64)
########################
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0111, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
41
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0361, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
42
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0272, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
43
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0239, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
44
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0149, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
45
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0323, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
46
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0218, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
47
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0243, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
48
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0217, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
49
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0164, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
50
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0142, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
51
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0401, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
52
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0328, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
53
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0102, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
54
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0491, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
55
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0106, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
56
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0426, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
57
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0382, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
58
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0246, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
59
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0191, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
60
########################
Validation Accuracy:
tensor(0.2883, dtype=torch.float64)
########################
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0476, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
61
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0393, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
62
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0086, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
63
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0090, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
64
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0669, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
65
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0267, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
66
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0152, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
67
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0144, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
68
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0135, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
69
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0345, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
70
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0198, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
71
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0235, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
72
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0437, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
73
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0224, grad_fn=<NllLossBackward>)
Epoch:
61
Batch:
74
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0139, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
0
########################
Validation Accuracy:
tensor(0.3067, dtype=torch.float64)
########################
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0133, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
1
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0183, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
2
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0173, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
3
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0136, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
4
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0195, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
5
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0200, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
6
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0258, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
7
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0364, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
8
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0414, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
9
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0179, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
10
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0273, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
11
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0254, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
12
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0283, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
13
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0357, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
14
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0147, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
15
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0363, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
16
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0421, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
17
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0217, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
18
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0242, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
19
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0204, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
20
########################
Validation Accuracy:
tensor(0.2883, dtype=torch.float64)
########################
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0236, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
21
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0194, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
22
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0156, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
23
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0325, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
24
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0283, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
25
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0251, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
26
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0114, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
27
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0316, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
28
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0251, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
29
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0119, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
30
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0215, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
31
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0206, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
32
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0244, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
33
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0097, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
34
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0235, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
35
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0459, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
36
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0442, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
37
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0173, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
38
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0478, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
39
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0444, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
40
########################
Validation Accuracy:
tensor(0.2800, dtype=torch.float64)
########################
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0198, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
41
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0227, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
42
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0147, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
43
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0471, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
44
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0150, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
45
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0309, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
46
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0675, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
47
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0501, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
48
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0331, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
49
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0230, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
50
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0310, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
51
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0364, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
52
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0234, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
53
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0405, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
54
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0120, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
55
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0118, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
56
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0275, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
57
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0257, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
58
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0373, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
59
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0257, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
60
########################
Validation Accuracy:
tensor(0.2800, dtype=torch.float64)
########################
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0141, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
61
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0080, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
62
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0353, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
63
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0155, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
64
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0282, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
65
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0168, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
66
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0338, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
67
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0148, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
68
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0150, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
69
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0154, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
70
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0089, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
71
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0230, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
72
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0262, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
73
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0215, grad_fn=<NllLossBackward>)
Epoch:
62
Batch:
74
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0196, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
0
########################
Validation Accuracy:
tensor(0.3100, dtype=torch.float64)
########################
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0271, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
1
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0229, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
2
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0200, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
3
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0156, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
4
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0216, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
5
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0232, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
6
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0316, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
7
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0111, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
8
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0059, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
9
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0378, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
10
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0095, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
11
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0121, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
12
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0233, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
13
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0284, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
14
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0142, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
15
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0275, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
16
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0137, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
17
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0275, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
18
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0297, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
19
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0373, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
20
########################
Validation Accuracy:
tensor(0.2800, dtype=torch.float64)
########################
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0305, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
21
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0199, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
22
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0066, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
23
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0225, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
24
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0436, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
25
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0385, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
26
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0322, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
27
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0143, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
28
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0243, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
29
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0203, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
30
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0491, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
31
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0177, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
32
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0266, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
33
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0349, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
34
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0232, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
35
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0305, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
36
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0469, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
37
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0337, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
38
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0320, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
39
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0310, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
40
########################
Validation Accuracy:
tensor(0.2833, dtype=torch.float64)
########################
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0561, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
41
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0417, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
42
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0195, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
43
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0117, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
44
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0355, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
45
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0311, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
46
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0212, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
47
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0180, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
48
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0514, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
49
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0278, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
50
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0279, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
51
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0553, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
52
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0230, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
53
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0241, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
54
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0273, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
55
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0209, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
56
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0311, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
57
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0464, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
58
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0272, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
59
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0198, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
60
########################
Validation Accuracy:
tensor(0.2950, dtype=torch.float64)
########################
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0273, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
61
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0560, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
62
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0208, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
63
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0315, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
64
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0315, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
65
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0363, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
66
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0305, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
67
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0325, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
68
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0535, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
69
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0444, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
70
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0327, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
71
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0249, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
72
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0113, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
73
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0244, grad_fn=<NllLossBackward>)
Epoch:
63
Batch:
74
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0180, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
0
########################
Validation Accuracy:
tensor(0.2883, dtype=torch.float64)
########################
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0153, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
1
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0244, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
2
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0259, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
3
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0079, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
4
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0148, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
5
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0145, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
6
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0251, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
7
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0317, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
8
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0111, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
9
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0170, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
10
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0416, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
11
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0108, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
12
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0134, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
13
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0247, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
14
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0315, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
15
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0213, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
16
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0252, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
17
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0322, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
18
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0151, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
19
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0161, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
20
########################
Validation Accuracy:
tensor(0.3050, dtype=torch.float64)
########################
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0217, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
21
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0314, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
22
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0131, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
23
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0080, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
24
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0299, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
25
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0110, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
26
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0291, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
27
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0252, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
28
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0232, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
29
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0071, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
30
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0260, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
31
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0244, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
32
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0206, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
33
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0210, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
34
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0404, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
35
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0174, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
36
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0186, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
37
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0272, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
38
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0308, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
39
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0212, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
40
########################
Validation Accuracy:
tensor(0.2950, dtype=torch.float64)
########################
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0198, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
41
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0298, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
42
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0164, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
43
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0107, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
44
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0172, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
45
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0115, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
46
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0123, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
47
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0254, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
48
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0276, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
49
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0184, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
50
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0105, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
51
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0321, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
52
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0469, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
53
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0306, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
54
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0260, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
55
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0342, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
56
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0220, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
57
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0690, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
58
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0413, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
59
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0646, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
60
########################
Validation Accuracy:
tensor(0.2867, dtype=torch.float64)
########################
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0392, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
61
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0280, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
62
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0228, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
63
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0590, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
64
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0377, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
65
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0445, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
66
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0249, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
67
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0461, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
68
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0230, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
69
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0260, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
70
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0247, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
71
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0356, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
72
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0306, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
73
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0576, grad_fn=<NllLossBackward>)
Epoch:
64
Batch:
74
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0131, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
0
########################
Validation Accuracy:
tensor(0.2867, dtype=torch.float64)
########################
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0592, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
1
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0273, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
2
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0291, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
3
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0421, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
4
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0178, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
5
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0313, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
6
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0215, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
7
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0200, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
8
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0100, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
9
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0193, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
10
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0227, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
11
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0365, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
12
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0259, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
13
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0137, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
14
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0237, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
15
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0337, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
16
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0141, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
17
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0179, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
18
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0825, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
19
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0152, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
20
########################
Validation Accuracy:
tensor(0.3150, dtype=torch.float64)
########################
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0231, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
21
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0665, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
22
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0232, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
23
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0202, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
24
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0222, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
25
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0114, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
26
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0230, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
27
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0274, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
28
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0320, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
29
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0389, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
30
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0141, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
31
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0589, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
32
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0148, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
33
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0113, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
34
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0198, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
35
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0246, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
36
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0204, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
37
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0470, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
38
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0123, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
39
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0442, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
40
########################
Validation Accuracy:
tensor(0.2833, dtype=torch.float64)
########################
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0462, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
41
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0283, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
42
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0299, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
43
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0197, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
44
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0176, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
45
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0254, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
46
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0283, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
47
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0220, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
48
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0282, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
49
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0442, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
50
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0221, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
51
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0279, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
52
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0255, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
53
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0393, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
54
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0304, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
55
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0206, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
56
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0405, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
57
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0184, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
58
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0327, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
59
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0381, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
60
########################
Validation Accuracy:
tensor(0.2950, dtype=torch.float64)
########################
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0198, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
61
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0216, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
62
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0297, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
63
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0347, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
64
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0431, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
65
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0178, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
66
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0175, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
67
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0305, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
68
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0253, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
69
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0142, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
70
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0210, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
71
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0188, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
72
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0101, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
73
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0125, grad_fn=<NllLossBackward>)
Epoch:
65
Batch:
74
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0194, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
0
########################
Validation Accuracy:
tensor(0.2733, dtype=torch.float64)
########################
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0374, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
1
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0169, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
2
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0181, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
3
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0145, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
4
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0079, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
5
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0382, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
6
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0223, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
7
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0401, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
8
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0327, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
9
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0220, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
10
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0393, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
11
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0243, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
12
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0206, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
13
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0248, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
14
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0204, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
15
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0350, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
16
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0317, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
17
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0191, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
18
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0225, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
19
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0509, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
20
########################
Validation Accuracy:
tensor(0.3017, dtype=torch.float64)
########################
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0152, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
21
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0104, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
22
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0264, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
23
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0380, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
24
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0317, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
25
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0194, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
26
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0179, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
27
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0185, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
28
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0249, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
29
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0424, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
30
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0288, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
31
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0271, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
32
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0290, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
33
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0478, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
34
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0084, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
35
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0539, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
36
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0306, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
37
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0195, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
38
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0190, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
39
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0139, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
40
########################
Validation Accuracy:
tensor(0.3100, dtype=torch.float64)
########################
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0304, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
41
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0124, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
42
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0229, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
43
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0280, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
44
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0337, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
45
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0384, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
46
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0173, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
47
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0435, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
48
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0347, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
49
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0614, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
50
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0180, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
51
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0189, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
52
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0536, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
53
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0593, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
54
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0174, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
55
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0215, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
56
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0087, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
57
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0280, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
58
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0559, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
59
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0443, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
60
########################
Validation Accuracy:
tensor(0.2817, dtype=torch.float64)
########################
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0296, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
61
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0148, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
62
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0307, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
63
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0058, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
64
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0274, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
65
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0457, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
66
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0334, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
67
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0304, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
68
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0286, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
69
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0453, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
70
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0181, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
71
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0200, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
72
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0143, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
73
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0424, grad_fn=<NllLossBackward>)
Epoch:
66
Batch:
74
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0275, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
0
########################
Validation Accuracy:
tensor(0.2900, dtype=torch.float64)
########################
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0056, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
1
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0104, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
2
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0170, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
3
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0109, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
4
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0229, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
5
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0151, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
6
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0113, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
7
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0242, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
8
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0163, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
9
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0055, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
10
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0104, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
11
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0273, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
12
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0163, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
13
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0142, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
14
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0209, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
15
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0227, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
16
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0464, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
17
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0185, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
18
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0084, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
19
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0214, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
20
########################
Validation Accuracy:
tensor(0.2933, dtype=torch.float64)
########################
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0380, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
21
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0280, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
22
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0193, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
23
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0068, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
24
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0088, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
25
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0229, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
26
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0120, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
27
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0219, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
28
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0081, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
29
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0269, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
30
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0323, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
31
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0258, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
32
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0356, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
33
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0165, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
34
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0121, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
35
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0311, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
36
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0084, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
37
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0153, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
38
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0166, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
39
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0213, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
40
########################
Validation Accuracy:
tensor(0.2883, dtype=torch.float64)
########################
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0258, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
41
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0225, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
42
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0296, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
43
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0079, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
44
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0208, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
45
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0132, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
46
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0156, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
47
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0067, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
48
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0132, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
49
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0228, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
50
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0167, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
51
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0332, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
52
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0319, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
53
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0415, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
54
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0170, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
55
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0188, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
56
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0194, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
57
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0223, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
58
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0158, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
59
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0450, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
60
########################
Validation Accuracy:
tensor(0.2917, dtype=torch.float64)
########################
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0145, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
61
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0328, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
62
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0266, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
63
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0315, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
64
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0199, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
65
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0206, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
66
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0154, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
67
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0288, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
68
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0215, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
69
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0188, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
70
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0167, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
71
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0113, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
72
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0502, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
73
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0081, grad_fn=<NllLossBackward>)
Epoch:
67
Batch:
74
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0118, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
0
########################
Validation Accuracy:
tensor(0.2917, dtype=torch.float64)
########################
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0445, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
1
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0076, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
2
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0113, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
3
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0145, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
4
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0213, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
5
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0434, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
6
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0109, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
7
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0132, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
8
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0280, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
9
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0291, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
10
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0238, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
11
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0071, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
12
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0198, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
13
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0288, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
14
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0133, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
15
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0078, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
16
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0624, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
17
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0186, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
18
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0270, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
19
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0784, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
20
########################
Validation Accuracy:
tensor(0.2967, dtype=torch.float64)
########################
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0273, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
21
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0246, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
22
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0099, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
23
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0111, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
24
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0175, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
25
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0200, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
26
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0179, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
27
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0258, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
28
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0227, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
29
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0197, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
30
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0263, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
31
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0103, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
32
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0503, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
33
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0148, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
34
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0125, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
35
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0414, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
36
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0248, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
37
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0438, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
38
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0138, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
39
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0093, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
40
########################
Validation Accuracy:
tensor(0.3067, dtype=torch.float64)
########################
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0299, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
41
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0234, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
42
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0302, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
43
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0098, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
44
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0181, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
45
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0109, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
46
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0184, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
47
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0135, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
48
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0755, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
49
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0234, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
50
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0150, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
51
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0080, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
52
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0139, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
53
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0383, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
54
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0477, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
55
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0160, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
56
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0173, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
57
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0633, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
58
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0394, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
59
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0291, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
60
########################
Validation Accuracy:
tensor(0.2933, dtype=torch.float64)
########################
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0196, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
61
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0378, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
62
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.0503, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
63
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0228, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
64
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0480, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
65
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0579, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
66
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0203, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
67
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0150, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
68
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0430, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
69
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0333, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
70
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0460, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
71
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0193, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
72
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0442, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
73
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0360, grad_fn=<NllLossBackward>)
Epoch:
68
Batch:
74
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0136, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
0
########################
Validation Accuracy:
tensor(0.2950, dtype=torch.float64)
########################
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0342, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
1
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0175, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
2
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0443, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
3
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0456, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
4
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0120, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
5
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0185, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
6
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0154, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
7
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0227, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
8
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0184, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
9
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0127, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
10
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0102, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
11
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0236, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
12
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0153, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
13
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0156, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
14
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0258, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
15
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0150, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
16
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0407, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
17
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0056, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
18
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0161, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
19
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0286, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
20
########################
Validation Accuracy:
tensor(0.3067, dtype=torch.float64)
########################
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0279, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
21
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0063, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
22
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0200, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
23
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0267, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
24
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0185, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
25
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0302, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
26
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0087, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
27
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0090, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
28
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0272, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
29
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0204, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
30
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0120, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
31
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0097, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
32
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0296, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
33
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0406, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
34
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0409, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
35
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0186, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
36
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0106, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
37
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0265, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
38
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0166, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
39
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0116, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
40
########################
Validation Accuracy:
tensor(0.3000, dtype=torch.float64)
########################
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0085, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
41
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0144, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
42
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0080, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
43
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0299, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
44
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0502, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
45
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0154, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
46
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0145, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
47
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0163, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
48
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0102, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
49
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0733, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
50
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0061, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
51
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0185, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
52
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0186, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
53
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0314, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
54
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0268, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
55
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0279, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
56
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0158, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
57
Accuracy:
tensor(0.9625, dtype=torch.float64)
Loss:
tensor(0.0711, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
58
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0114, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
59
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0215, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
60
########################
Validation Accuracy:
tensor(0.2783, dtype=torch.float64)
########################
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0329, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
61
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0632, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
62
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0175, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
63
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0239, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
64
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0523, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
65
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0612, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
66
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0106, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
67
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0357, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
68
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0253, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
69
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0160, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
70
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0354, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
71
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0156, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
72
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0269, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
73
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0284, grad_fn=<NllLossBackward>)
Epoch:
69
Batch:
74
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0197, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
0
########################
Validation Accuracy:
tensor(0.2950, dtype=torch.float64)
########################
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0216, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
1
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0102, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
2
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0239, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
3
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0238, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
4
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0287, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
5
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0347, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
6
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0194, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
7
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0139, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
8
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0340, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
9
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0315, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
10
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0249, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
11
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0160, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
12
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0598, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
13
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0110, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
14
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0140, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
15
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0234, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
16
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0104, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
17
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0424, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
18
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0059, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
19
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0210, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
20
########################
Validation Accuracy:
tensor(0.2800, dtype=torch.float64)
########################
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0037, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
21
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0218, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
22
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0126, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
23
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0278, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
24
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0202, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
25
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0131, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
26
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0181, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
27
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0150, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
28
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0185, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
29
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0189, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
30
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0186, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
31
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0120, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
32
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0206, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
33
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0620, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
34
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0344, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
35
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0301, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
36
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0231, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
37
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0269, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
38
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0172, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
39
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0668, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
40
########################
Validation Accuracy:
tensor(0.2967, dtype=torch.float64)
########################
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0348, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
41
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0239, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
42
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0348, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
43
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0263, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
44
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0242, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
45
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0140, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
46
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0493, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
47
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0511, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
48
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0266, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
49
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0162, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
50
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0206, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
51
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0297, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
52
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0549, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
53
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0267, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
54
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0164, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
55
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0241, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
56
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0182, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
57
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0068, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
58
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0156, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
59
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0285, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
60
########################
Validation Accuracy:
tensor(0.2917, dtype=torch.float64)
########################
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0153, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
61
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0213, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
62
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0302, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
63
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0192, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
64
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0353, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
65
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0094, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
66
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0416, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
67
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0181, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
68
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0184, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
69
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0487, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
70
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0141, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
71
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0328, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
72
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0319, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
73
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0409, grad_fn=<NllLossBackward>)
Epoch:
70
Batch:
74
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0150, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
0
########################
Validation Accuracy:
tensor(0.3100, dtype=torch.float64)
########################
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0245, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
1
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0180, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
2
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0576, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
3
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0151, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
4
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0419, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
5
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0329, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
6
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0219, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
7
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0187, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
8
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0582, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
9
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0331, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
10
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0167, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
11
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0753, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
12
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0557, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
13
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0239, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
14
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.0721, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
15
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0318, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
16
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0310, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
17
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0137, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
18
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0568, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
19
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0305, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
20
########################
Validation Accuracy:
tensor(0.2900, dtype=torch.float64)
########################
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0083, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
21
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0296, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
22
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0376, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
23
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0323, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
24
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0218, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
25
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0416, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
26
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0317, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
27
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0134, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
28
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0251, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
29
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0217, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
30
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0163, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
31
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0129, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
32
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0136, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
33
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0315, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
34
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0315, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
35
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0212, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
36
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0557, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
37
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0217, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
38
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0159, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
39
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0158, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
40
########################
Validation Accuracy:
tensor(0.2917, dtype=torch.float64)
########################
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0578, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
41
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0313, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
42
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0559, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
43
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0357, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
44
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0261, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
45
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0202, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
46
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0220, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
47
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0347, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
48
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0291, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
49
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0514, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
50
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0140, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
51
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0264, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
52
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0257, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
53
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0306, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
54
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0319, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
55
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0481, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
56
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0236, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
57
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0252, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
58
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0265, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
59
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0503, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
60
########################
Validation Accuracy:
tensor(0.2967, dtype=torch.float64)
########################
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0136, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
61
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0223, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
62
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0187, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
63
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0350, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
64
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0529, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
65
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0340, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
66
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0175, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
67
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0463, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
68
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0560, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
69
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0477, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
70
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0455, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
71
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0256, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
72
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0113, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
73
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0585, grad_fn=<NllLossBackward>)
Epoch:
71
Batch:
74
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0231, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
0
########################
Validation Accuracy:
tensor(0.2950, dtype=torch.float64)
########################
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0208, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
1
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0271, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
2
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0198, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
3
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0265, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
4
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0285, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
5
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0120, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
6
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0108, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
7
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0340, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
8
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0350, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
9
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0335, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
10
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0167, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
11
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0182, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
12
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0271, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
13
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0365, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
14
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0182, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
15
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0210, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
16
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0283, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
17
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0192, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
18
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0335, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
19
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0143, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
20
########################
Validation Accuracy:
tensor(0.3033, dtype=torch.float64)
########################
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0209, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
21
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0243, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
22
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0440, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
23
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0058, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
24
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0209, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
25
Accuracy:
tensor(0.9583, dtype=torch.float64)
Loss:
tensor(0.1135, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
26
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0132, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
27
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0358, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
28
Accuracy:
tensor(0.9375, dtype=torch.float64)
Loss:
tensor(0.1473, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
29
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0213, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
30
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0464, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
31
Accuracy:
tensor(0.9375, dtype=torch.float64)
Loss:
tensor(0.1791, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
32
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0204, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
33
Accuracy:
tensor(0.9417, dtype=torch.float64)
Loss:
tensor(0.2272, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
34
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0933, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
35
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0350, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
36
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0248, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
37
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0721, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
38
Accuracy:
tensor(0.9458, dtype=torch.float64)
Loss:
tensor(0.1513, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
39
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0412, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
40
########################
Validation Accuracy:
tensor(0.2950, dtype=torch.float64)
########################
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0495, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
41
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.0816, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
42
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0552, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
43
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0482, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
44
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0251, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
45
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0347, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
46
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0435, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
47
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0557, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
48
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0411, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
49
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0614, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
50
Accuracy:
tensor(0.9542, dtype=torch.float64)
Loss:
tensor(0.1605, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
51
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0532, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
52
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0282, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
53
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0799, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
54
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0486, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
55
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0740, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
56
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0235, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
57
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0555, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
58
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0169, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
59
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0631, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
60
########################
Validation Accuracy:
tensor(0.2517, dtype=torch.float64)
########################
Accuracy:
tensor(0.9583, dtype=torch.float64)
Loss:
tensor(0.0930, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
61
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0433, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
62
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0774, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
63
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0452, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
64
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.1120, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
65
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0378, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
66
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0623, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
67
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0351, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
68
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0635, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
69
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0475, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
70
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0437, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
71
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0450, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
72
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0436, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
73
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0156, grad_fn=<NllLossBackward>)
Epoch:
72
Batch:
74
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0612, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
0
########################
Validation Accuracy:
tensor(0.2783, dtype=torch.float64)
########################
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0236, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
1
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0318, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
2
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0272, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
3
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0297, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
4
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0156, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
5
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0308, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
6
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0383, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
7
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0325, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
8
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0300, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
9
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0169, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
10
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0472, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
11
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0266, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
12
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0215, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
13
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0231, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
14
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0297, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
15
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0208, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
16
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0326, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
17
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0187, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
18
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0265, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
19
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0168, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
20
########################
Validation Accuracy:
tensor(0.2817, dtype=torch.float64)
########################
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0133, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
21
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0429, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
22
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0063, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
23
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0201, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
24
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0251, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
25
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0221, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
26
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0606, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
27
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0349, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
28
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0240, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
29
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0431, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
30
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0251, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
31
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0248, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
32
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0327, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
33
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0343, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
34
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0126, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
35
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0674, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
36
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0115, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
37
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0350, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
38
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0359, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
39
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0310, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
40
########################
Validation Accuracy:
tensor(0.3000, dtype=torch.float64)
########################
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0998, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
41
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0150, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
42
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0199, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
43
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0395, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
44
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0286, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
45
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0285, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
46
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0376, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
47
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0213, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
48
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0160, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
49
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0338, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
50
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0146, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
51
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0480, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
52
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0076, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
53
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0107, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
54
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0187, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
55
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0083, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
56
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0216, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
57
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0305, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
58
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0495, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
59
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0387, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
60
########################
Validation Accuracy:
tensor(0.2850, dtype=torch.float64)
########################
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0200, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
61
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0352, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
62
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0313, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
63
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0168, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
64
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0203, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
65
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0303, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
66
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0154, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
67
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0233, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
68
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0364, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
69
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0428, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
70
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0219, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
71
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0303, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
72
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0406, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
73
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0179, grad_fn=<NllLossBackward>)
Epoch:
73
Batch:
74
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0354, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
0
########################
Validation Accuracy:
tensor(0.3017, dtype=torch.float64)
########################
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0353, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
1
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0344, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
2
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0264, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
3
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0121, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
4
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0219, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
5
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0128, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
6
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0219, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
7
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0186, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
8
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0222, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
9
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0118, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
10
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0219, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
11
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0735, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
12
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0258, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
13
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0134, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
14
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0192, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
15
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0147, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
16
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0156, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
17
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0324, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
18
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0203, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
19
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0165, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
20
########################
Validation Accuracy:
tensor(0.3017, dtype=torch.float64)
########################
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0134, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
21
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0315, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
22
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0380, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
23
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0327, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
24
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0196, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
25
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0149, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
26
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0528, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
27
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0163, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
28
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0219, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
29
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0064, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
30
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0323, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
31
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0718, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
32
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0265, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
33
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0357, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
34
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0108, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
35
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0208, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
36
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0429, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
37
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0196, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
38
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0425, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
39
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0265, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
40
########################
Validation Accuracy:
tensor(0.2900, dtype=torch.float64)
########################
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0198, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
41
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0218, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
42
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0222, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
43
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0362, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
44
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0115, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
45
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0222, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
46
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0157, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
47
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0227, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
48
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0109, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
49
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0198, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
50
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0088, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
51
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0145, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
52
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0143, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
53
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0511, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
54
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0585, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
55
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0281, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
56
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0159, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
57
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0558, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
58
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0138, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
59
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0086, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
60
########################
Validation Accuracy:
tensor(0.2783, dtype=torch.float64)
########################
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0118, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
61
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0200, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
62
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0124, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
63
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0376, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
64
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0426, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
65
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0172, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
66
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0268, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
67
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0149, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
68
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0254, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
69
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0395, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
70
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0464, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
71
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0199, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
72
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0465, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
73
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0248, grad_fn=<NllLossBackward>)
Epoch:
74
Batch:
74
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0144, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
0
########################
Validation Accuracy:
tensor(0.2850, dtype=torch.float64)
########################
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0363, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
1
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0161, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
2
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0212, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
3
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0609, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
4
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0073, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
5
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0192, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
6
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0248, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
7
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0372, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
8
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0053, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
9
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0204, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
10
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0207, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
11
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0080, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
12
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0329, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
13
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0124, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
14
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0072, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
15
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0222, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
16
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0196, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
17
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0294, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
18
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0105, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
19
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0342, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
20
########################
Validation Accuracy:
tensor(0.3050, dtype=torch.float64)
########################
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0276, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
21
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0275, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
22
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0356, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
23
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0192, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
24
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0054, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
25
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0085, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
26
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0160, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
27
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0295, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
28
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0259, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
29
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0155, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
30
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0569, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
31
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0402, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
32
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0242, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
33
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0069, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
34
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0054, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
35
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0351, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
36
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0515, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
37
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0232, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
38
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0116, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
39
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0221, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
40
########################
Validation Accuracy:
tensor(0.2917, dtype=torch.float64)
########################
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0219, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
41
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0639, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
42
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0238, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
43
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0237, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
44
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0153, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
45
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0163, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
46
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0412, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
47
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0508, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
48
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0193, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
49
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0429, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
50
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0151, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
51
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0272, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
52
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0118, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
53
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0169, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
54
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0549, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
55
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0269, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
56
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0289, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
57
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0160, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
58
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0215, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
59
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0194, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
60
########################
Validation Accuracy:
tensor(0.2867, dtype=torch.float64)
########################
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0178, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
61
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0409, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
62
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0183, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
63
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0311, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
64
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0089, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
65
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0366, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
66
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0124, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
67
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0355, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
68
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0261, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
69
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0330, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
70
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0171, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
71
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0103, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
72
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0394, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
73
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0302, grad_fn=<NllLossBackward>)
Epoch:
75
Batch:
74
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0299, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
0
########################
Validation Accuracy:
tensor(0.2950, dtype=torch.float64)
########################
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0042, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
1
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0065, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
2
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0117, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
3
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0312, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
4
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0344, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
5
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0270, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
6
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0296, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
7
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0265, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
8
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0082, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
9
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0272, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
10
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0109, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
11
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0367, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
12
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0328, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
13
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0217, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
14
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0189, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
15
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0372, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
16
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0326, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
17
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0340, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
18
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0107, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
19
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0219, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
20
########################
Validation Accuracy:
tensor(0.2950, dtype=torch.float64)
########################
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0134, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
21
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0038, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
22
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0164, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
23
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0364, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
24
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0289, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
25
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0211, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
26
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0281, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
27
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0215, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
28
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0147, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
29
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0536, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
30
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0138, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
31
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0260, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
32
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0234, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
33
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0295, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
34
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0079, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
35
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0178, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
36
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0406, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
37
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0285, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
38
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0187, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
39
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0210, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
40
########################
Validation Accuracy:
tensor(0.2950, dtype=torch.float64)
########################
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0060, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
41
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0153, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
42
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0779, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
43
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0240, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
44
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0154, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
45
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0287, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
46
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0120, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
47
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0183, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
48
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0075, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
49
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0161, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
50
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0402, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
51
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0089, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
52
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0148, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
53
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0223, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
54
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0462, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
55
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0487, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
56
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0051, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
57
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0243, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
58
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0154, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
59
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0418, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
60
########################
Validation Accuracy:
tensor(0.2883, dtype=torch.float64)
########################
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0279, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
61
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0179, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
62
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0252, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
63
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0184, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
64
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0231, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
65
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0294, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
66
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0197, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
67
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0410, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
68
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0318, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
69
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0148, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
70
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0073, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
71
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0106, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
72
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0226, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
73
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0107, grad_fn=<NllLossBackward>)
Epoch:
76
Batch:
74
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0162, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
0
########################
Validation Accuracy:
tensor(0.2867, dtype=torch.float64)
########################
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0185, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
1
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0295, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
2
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0071, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
3
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0162, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
4
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0206, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
5
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0132, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
6
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0140, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
7
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0191, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
8
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0255, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
9
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0171, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
10
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0287, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
11
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0075, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
12
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0226, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
13
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0242, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
14
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0249, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
15
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0347, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
16
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0215, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
17
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0167, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
18
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0101, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
19
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0056, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
20
########################
Validation Accuracy:
tensor(0.2917, dtype=torch.float64)
########################
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0117, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
21
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0410, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
22
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0049, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
23
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0112, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
24
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0223, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
25
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0218, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
26
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0153, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
27
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0083, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
28
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0133, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
29
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0135, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
30
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0143, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
31
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0188, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
32
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0251, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
33
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0309, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
34
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0092, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
35
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0197, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
36
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0229, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
37
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0185, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
38
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0264, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
39
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0200, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
40
########################
Validation Accuracy:
tensor(0.3000, dtype=torch.float64)
########################
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0302, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
41
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0142, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
42
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0452, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
43
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0073, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
44
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0181, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
45
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0094, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
46
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0235, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
47
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0198, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
48
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0355, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
49
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0189, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
50
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0302, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
51
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0264, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
52
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0178, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
53
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0207, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
54
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0121, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
55
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0231, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
56
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0348, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
57
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0458, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
58
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0444, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
59
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0201, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
60
########################
Validation Accuracy:
tensor(0.3050, dtype=torch.float64)
########################
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0155, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
61
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0509, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
62
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.0765, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
63
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0119, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
64
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0323, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
65
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0451, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
66
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0383, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
67
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0556, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
68
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0299, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
69
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0101, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
70
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0264, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
71
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0335, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
72
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0538, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
73
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0355, grad_fn=<NllLossBackward>)
Epoch:
77
Batch:
74
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0085, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
0
########################
Validation Accuracy:
tensor(0.2833, dtype=torch.float64)
########################
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0170, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
1
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0439, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
2
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0519, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
3
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0103, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
4
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0192, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
5
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0115, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
6
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0109, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
7
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0393, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
8
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0333, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
9
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0104, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
10
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0133, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
11
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0626, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
12
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0198, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
13
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0176, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
14
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0121, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
15
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0368, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
16
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0355, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
17
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0334, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
18
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0264, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
19
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0188, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
20
########################
Validation Accuracy:
tensor(0.2950, dtype=torch.float64)
########################
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0378, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
21
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0138, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
22
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0589, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
23
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0893, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
24
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0381, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
25
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0112, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
26
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0370, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
27
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0306, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
28
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0570, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
29
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0233, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
30
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0264, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
31
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.0840, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
32
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0106, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
33
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0652, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
34
Accuracy:
tensor(0.9458, dtype=torch.float64)
Loss:
tensor(0.1967, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
35
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0659, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
36
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0775, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
37
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0552, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
38
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0435, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
39
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0180, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
40
########################
Validation Accuracy:
tensor(0.2983, dtype=torch.float64)
########################
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0266, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
41
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0725, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
42
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0193, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
43
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0612, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
44
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0146, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
45
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0145, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
46
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0344, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
47
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0598, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
48
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0727, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
49
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0149, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
50
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0307, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
51
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0269, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
52
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0173, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
53
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0272, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
54
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0299, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
55
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0150, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
56
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0479, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
57
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0331, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
58
Accuracy:
tensor(0.9750, dtype=torch.float64)
Loss:
tensor(0.0644, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
59
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0373, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
60
########################
Validation Accuracy:
tensor(0.2850, dtype=torch.float64)
########################
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0099, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
61
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0342, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
62
Accuracy:
tensor(0.9708, dtype=torch.float64)
Loss:
tensor(0.0818, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
63
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0249, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
64
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0045, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
65
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0408, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
66
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0236, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
67
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0493, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
68
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0202, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
69
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0323, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
70
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0131, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
71
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0431, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
72
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0231, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
73
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0165, grad_fn=<NllLossBackward>)
Epoch:
78
Batch:
74
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0213, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
0
########################
Validation Accuracy:
tensor(0.2967, dtype=torch.float64)
########################
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0237, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
1
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0132, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
2
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0379, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
3
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0211, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
4
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0215, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
5
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0235, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
6
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0099, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
7
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0253, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
8
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0282, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
9
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0350, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
10
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0340, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
11
Accuracy:
tensor(0.9667, dtype=torch.float64)
Loss:
tensor(0.0921, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
12
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0170, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
13
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0097, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
14
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0253, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
15
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0426, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
16
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0162, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
17
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0052, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
18
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0393, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
19
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0121, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
20
########################
Validation Accuracy:
tensor(0.3033, dtype=torch.float64)
########################
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0046, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
21
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0253, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
22
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0180, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
23
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0358, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
24
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0199, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
25
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0263, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
26
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0334, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
27
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0088, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
28
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0071, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
29
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0415, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
30
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0122, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
31
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0453, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
32
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0250, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
33
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0091, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
34
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0188, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
35
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0296, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
36
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0151, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
37
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0120, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
38
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0237, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
39
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0518, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
40
########################
Validation Accuracy:
tensor(0.2850, dtype=torch.float64)
########################
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0177, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
41
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0208, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
42
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0180, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
43
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0144, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
44
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0252, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
45
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0138, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
46
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0221, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
47
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0260, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
48
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0210, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
49
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0230, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
50
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0287, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
51
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0090, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
52
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0237, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
53
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0191, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
54
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0177, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
55
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0152, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
56
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0274, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
57
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0128, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
58
Accuracy:
tensor(0.9792, dtype=torch.float64)
Loss:
tensor(0.0409, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
59
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0424, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
60
########################
Validation Accuracy:
tensor(0.3100, dtype=torch.float64)
########################
Accuracy:
tensor(1., dtype=torch.float64)
Loss:
tensor(0.0144, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
61
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0277, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
62
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0320, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
63
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0146, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
64
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0117, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
65
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0542, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
66
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0279, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
67
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0148, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
68
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0120, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
69
Accuracy:
tensor(0.9917, dtype=torch.float64)
Loss:
tensor(0.0215, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
70
Accuracy:
tensor(0.9833, dtype=torch.float64)
Loss:
tensor(0.0302, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
71
Accuracy:
tensor(0.9958, dtype=torch.float64)
Loss:
tensor(0.0161, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
72
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0246, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
73
Accuracy:
tensor(0.9875, dtype=torch.float64)
Loss:
tensor(0.0376, grad_fn=<NllLossBackward>)
Epoch:
79
Batch:
74

Plotting the Training Accuracy, Validation Accuracy, and Training Loss¶

plt.figure(1, figsize=(12, 8))
plt.plot(trainacc, '-', color='orange',linewidth=2, label='Train Accuracy')
plt.title('Train Accuracy')
plt.ylabel('Accuracy')
plt.legend(loc='best')
plt.grid()
plt.show()


plt.figure(2, figsize=(12, 8))
plt.plot(testacc, '-', color='blue',linewidth=2, label='Test Accuracy')
plt.title('Test Accuracy')
plt.ylabel('Accuracy')
plt.legend(loc='best')
plt.grid()
plt.show()

plt.figure(3, figsize=(12, 8))
plt.plot(trainloss, '-', color='green',linewidth=2, label='Training Loss')
plt.title('Training Loss')
plt.ylabel('Loss')
plt.legend(loc='best')
plt.grid()
plt.show()

Predictions¶

Generating predictions for the testing set and preparing for submission.

prob = output.tolist()
predictions = np.argmax(prob, axis=1)

pred_label = []
for i in predictions:
    pred_label.append(classes[i])
from numpy import asarray
from numpy import savetxt
import pandas as pd                                        
# define id
id = np.arange(0,20000)
labels=np.asarray(pred_label)
labels.reshape(20000,1)
print(labels.shape)
d = {'Category':labels}
df = pd.DataFrame(data=d)
df.to_csv('submission.csv')

(20000,)